p a s r e p o r t 592

PLANNING FOR
AUTONOMOUS
MOBILITY
Jeremy Crute, William Riggs, aicp, Timothy S. Chapin, and Lindsay Stevens, aicp
 ABOUT THE AUTHORS
APA RESEARCH MISSION
Jeremy Crute has been the senior planner in Florida State University’s Depart-
ment of Urban and Regional Planning for the last four years. There he has man-
APA conducts applied, policy-relevant research that
aged a wide range of applied and scholarly research projects on community
advances the state of the art in planning practice. redevelopment, transportation, and land-use issues. He has prior professional
APA’s National Centers for Planning—the Green community development experience in Chattanooga, Tennessee, and Atlanta,
Communities Center, the Hazards Planning Center, Georgia. He holds a master of science degree in urban and regional planning
and the Planning and Community Health Center— from Florida State University, and bachelors degrees in economics and com-
guide and advance a research directive that address- munity development from Covenant College.
es important societal issues. APA’s research, educa-
William (Billy) Riggs, aicp, phd, leed ap, is a global expert and thought leader
tion, and advocacy programs help planners create in the areas of future mobility and smart transportation, housing, economics,
communities of lasting value by developing and dis- and urban development. He is a professor at the University of San Francisco
seminating information, tools, and applications for School of Management, and an advisor to multiple companies and start-ups on
built and natural environments. technology, smart mobility, and urban development. This follows two decades
of work as a planner, economist, and engineer. He has written more than 100
publications; been featured in multiple global media outlets such as the Econo-
mist, Wall Street Journal, Washington Post, and The Atlantic; and is the author of
Disruptive Transport: Driverless Cars, Transport Innovation and the Sustainable
Since 1949, the American Planning Association’s Planning Advisory Service has City of Tomorrow (Routledge, forthcoming). He sits on Palo Alto, California’s
provided planners with expert research, authoritative information on best prac- Planning and Transportation Commission.
tices and innovative solutions, and practical tools to help them manage on-the-
Timothy S. Chapin, phd, is the dean of the College of Social Sciences and Pub-
job challenges. PAS Reports are produced in the Research Department of APA.
lic Policy and a professor in the Department of Urban & Regional Planning
James M. Drinan, jd, Chief Executive Officer; David C. Rouse, faicp, Managing
at Florida State University, where he has studied the effectiveness of Florida’s
Director of Research and Advisory Services; Ann F. Dillemuth, aicp, Editor. APA growth management regime, assessed the success of planning efforts in shaping
membership includes access to all PAS publications, including PAS Reports, PAS development outcomes, and evaluated the role of sports facilities in the pro-
Memo, and PAS QuickNotes. Learn more at planning.org/pas. motion of urban redevelopment. His current research interests revolve around
how autonomous vehicles and national demographic trends will influence ur-
ban development and transportation activities in Florida and the United States.
Recipients of print editions of PAS Reports with missing or damaged copies:
Over his career, he has secured more than $3 million in funding from federal,
Contact APA Customer Service 312-431-9100 or [email protected]) state, and local governments to support his research. Chapin has also served as
within 90 days of the publication date. interim editor, senior associate editor, and review editor for the Journal of the
American Planning Association.
©September 2018 American Planning Association, which has offices at 205
N. Michigan Ave., Suite 1200, Chicago, IL 60601-5927, and 1030 15th St., NW, Lindsay Stevens, aicp, is the land program manager for the Florida Chapter
of The Nature Conservancy, where she provides strategic leadership for land
Suite 750 West, Washington, DC 20005-1503; planning.org. All rights reserved.
protection efforts including implementation of conservation real estate transac-
No part of this publication may be reproduced or utilized in any form or by any tions and planning for the protection of critical areas in Florida. Prior to joining
means without permission in writing from APA. The Nature Conservancy, Stevens was planner-in-residence for the Florida State
University Department of Urban and Regional Planning, where she worked
on a wide variety of land-use and transportation planning projects, includ-
ISBN: 978-1-61190-200-6 ing the potential impact of automated vehicles on development of urban areas
in Florida. Stevens has also worked for Wakulla County, Florida, as assistant
county administrator, with The Trust for Public Land on conservation real es-
Email: [email protected] tate and planning projects, and as a land-use and real estate attorney with Baker
& Hostetler LLP.

ON THE COVER

Preparing cities for autonomous (self-driving) transportation re-

search project, Chicago’s West Loop (Stantec’s Urban Places)
TABLE OF CONTENTS

INFOGRAPHIC 2

EXECUTIVE SUMMARY 3

CHAPTER 1 INTRODUCTION 6
Planning, Uncertainty, and the Revolution 8
About This Report 9

CHAPTER 2 AUTONOMOUS VEHICLES 101 14

The Technology: What Is It and How Does It Work? 15
AV Ownership and the Rise of Shared AVs 20
Timeline for AV Adoption 21
Conclusion 24

CHAPTER 3 PLANNING OPPORTUNITIES AND CHALLENGES IN AN AV WORLD 26

AV-Related Planning Opportunities 27
AV-Related Planning Challenges 32
Other Secondary Impacts of AVs 34
Conclusion 39

CHAPTER 4 POTENTIAL IMPACTS OF AVS ON THE BUILT ENVIRONMENT 40

Rights-of-Way 41
Access Management 43
Signage and Signalization 43
Interface With Bicycles and Pedestrians 45
Parking 46
Redevelopment Opportunities 49
New Design Paradigms 51
Conclusion 51

CHAPTER 5 CONSIDERATIONS FOR POLICY MAKING AND INFRASTRUCTURE INVESTMENTS 52

Addressing Planning Opportunities and Challenges 53
Transitioning to an AV-Dominated System 67
Addressing AV in Community Planning Processes 70
Conclusion 71

CHAPTER 6 CONCLUSION: THE TRANSFORMATIVE IMPACTS OF AVS 72

REFERENCES 77

ACKNOWLEDGMENTS 84
 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, I N F O G R A P H I C

AVs:
Coming to a community near you
A future with autonomous vehicles (AVs) is
closer than you may think. PAS Report 592,
Planning for Autonomous Mobility,
will help you prepare for the
implications and changes.

THE FUTURE IS HERE BIG CHANGES MORE PLANNING IS NEEDED

  
The earliest year an
automaker projects
1 shared AV policies
2020 they will have a fully AV in action
autonomous car Of the 500 largest US cities,
available for how many of them have
the public AV policies?
could replace
between 9 and 11
privately-owned
vehicles.

2040 43 %
or more:
expected
reduction
rate in vehicles
Estimated privately owned
year when

50 %
of cars will
be AVs Parking,
automotive
uses, transit, ■ No AV policy: 95%
biking, and
■ AV in the comprehensive
walking will
plan: 3%
all be affected
by this shift ■ AV ordinance: 2%
on our roads. Source: Riggs, Steins, and Chavan 2018

2 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, E X E C U T I V E S U M M A RY

EXECUTIVE SUMMARY

PAS Report 592, Planning for Autonomous Mobility, serves as a call to action for professional planners, especially those
working in the public sector in the transportation and land-use arenas. Autonomous vehicles (AVs) will transform the built
environment in the coming decades, and communities must begin planning for AVs now to ensure that this new technol-
ogy is harnessed in beneficial ways. The primary goals of this PAS Report are to (1) provide planners and policy makers
with the foundational knowledge necessary to anticipate potential impacts of AVs on communities, and (2) support and
spur development of policy solutions and infrastructure investments that ensure attractive, people-friendly, equitable, and
safe urban environments.

This PAS Report explores the many benefits that AVs However, this transformative potential does not come
may provide, but also looks at the challenges that AVs will solely from AV technology. The convergence of technological
bring to communities. The many potential impacts of AVs in- advances with the rise of the shared economy and ride-sharing
clude the following: services like Lyft and Uber could transform the predominant
• altering the design of rights-of-way mobility paradigm from privately owned to shared vehicles. A
• changing access management practices transition to shared mobility would have significant implica-
• influencing the form and function of traffic signage tions for the size of the vehicle fleet, traffic congestion, parking,
and signalization and urban design. Ultimately, shared AVs could lead to a much
• bringing massive changes to pedestrian and bicycle smaller vehicle fleet as each vehicle completes more trips.
networks The timeline for AVs’ arrival will help determine how
• reducing the demand and altering the design and planners need to respond, but predicting AV adoption rates
location of parking is a difficult task. The technology is advancing rapidly and
• creating redevelopment opportunities in urban and several companies anticipate having AVs available for sale in
suburban locales the early 2020s. The novelty and convenience of autonomous
driving could speed adoption, and highly, if not fully, autono-
It is imperative for planners to begin considering how mous vehicles could easily represent at least a quarter of the
AVs will affect our built environment and how this technol- vehicles on the road in less than 15 years. However, vehicle
ogy can contribute to community livability, efficient trans- prices, regulatory delays, and uncertainties surrounding in-
portation systems, and vibrant public spaces. surance, legal liability, testing and validation procedures, and
cybersecurity could delay AVs’ market availability.
As AVs take on a greater share of the vehicle fleet over
WHAT ARE AVS AND WHEN time, there will be a complex and messy transition period
WILL THEY GET HERE? where autonomous and human-driven vehicles share the
road. Regardless of the exact timeline, AVs are coming, and
Autonomous vehicles encompass a wide range of emerging they will irrevocably change transportation systems, the built
technologies that had previously been the stuff of science fic- environment, and our communities.
tion. Already, advanced driver assistance systems are improv-
ing safety by controlling specific driving functions; fully au-
tonomous vehicles will be capable of driving without human HOW WILL AVS CHANGE TRANSPORTATION
operation. Connected vehicle technologies will enable these AND THE BUILT ENVIRONMENT?
vehicles to communicate and coordinate amongst themselves
and the surrounding infrastructure, further improving travel Like the changes to society already brought by shared mo-
safety and efficiency. Advancements in electric vehicle and bility and digital ride-hailing services, AVs will disrupt the
traffic management systems will complement the emergence way that citizens travel and businesses operate. The technol-
of AVs and magnify the benefits they promise to provide. ogy brings both promise and peril.

www.planning.org AMERICAN PLANNING ASSOCIATION 3

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, E X E C U T I V E S U M M A RY

This PAS Report explains how AVs have the potential to and productive landscapes, encouraging a diversity of hous-
improve the safety and efficiency of transportation systems, ing types and densities, promoting a range of travel modes,
reduce vehicle emissions, and improve the mobility of trans- and serving the built environment with quality infrastruc-
portation-disadvantaged populations. However, AVs will not ture in core urban and suburban areas. Planners must pre-
solve all planning problems and will create new ones, such as pare their communities for the wide range of possible impacts
the need for drop-off zones, vehicle storage or circulators for that AV technology may have on transportation and the built
vehicles as they await users, and expensive new transporta- environment, integrating AV considerations into community
tion infrastructure to maximize the benefits of the technol- planning practices through techniques such as visioning and
ogy. AVs may also reinforce urban sprawl by reducing the scenario planning.
monetary and perceived costs of travel, further decreasing This report offers important recommendations for ac-
the friction of distance for households and businesses. In ad- tion. First, parking standards and requirements require im-
dition, AVs will have important ramifications on several oth- mediate attention, as parking demand and need is changing
er key planning areas, including transit, public health, and with each passing day. While this has been the case for a
social equity. In each of these areas, proactive steps will be long while, it is of heightened importance in an autonomous
required for planners and policy makers to capitalize on the future. Second, cities must bolster transportation demand
opportunities while mitigating the challenges. management efforts and link those more overtly to a shared
Two of the most significant areas this report identifies and autonomous mobility future. These efforts can be en-
in the planning realm are (1) parking and (2) the curb. Plan- hanced by curbside pick-up and drop-off policies, with this
ners and policy makers have seen significant changes in re- report illustrating some visions for how that can occur.
cent years in parking inventory and curbside management. Third, planners must rethink the right-of-way for al-
These shifts to reduced parking demands and ride sharing ternative modes, recognizing that AVs offer an opportunity
with curbside pick-up and drop-off will only become more to “right-size” roads at the human scale. Building upon the
pronounced with the rise of AVs. complete streets movement, in the longer run AVs offer the
AVs will also affect the built environment in a myriad of potential for aggressive road diets that reallocate space pre-
ways, including new right-of-way designs, changes to access viously used for automobiles back to human-powered and
management practices, reconsideration of signage and sig- active travel modes. Communities should prioritize these
nalization, new models for pedestrian and bicycle networks, modes in their comprehensive and general plans and begin
and reductions in demand and changes to the location of to experiment with new roadway typologies that provide ac-
parking. The report also forecasts redevelopment opportuni- commodation to these modes. Right-of-way reallocation also
ties in urban and suburban locales from former auto-serving holds the potential to provide new space for green infrastruc-
uses, while narrower rights-of-way, enhanced bicycle and ture, public gathering places, and other features that can help
pedestrian facilities, and redevelopment may create excellent achieve various community goals.
opportunities to revitalize urban centers. Finally, communities should pursue the opportunities
However, by making travel easier and more convenient, that AVs bring to improving transit service. Transit plan-
AVs could undermine these opportunities by encouraging ners must welcome this change and seek opportunities to
sprawl, expanding the already voracious metropolitan de- pilot transit-specific applications of AV technology. Nu-
velopment that exists in the United States. Careful planning merous jurisdictions are piloting or implementing fully
and policy will be required to shape these built environment autonomous transit routes on public roads across the coun-
impacts to ensure the creation of vibrant, sustainable, and re- try. These efforts showcase the power of AV technology to
silient communities. provide transit services that provide accessibility to under-
served portions of communities.

WHAT CAN PLANNERS DO NOW?

CONCLUSION
The key takeaway from this PAS Report is this: There is no
substitute for quality, comprehensive, long-term oriented The private automobile has played a major role in shaping
planning. It is imperative that the profession remain commit- the built form of cities and suburbs. In almost all commu-
ted to its foundational principles of protecting sensitive land nities, development and land-use patterns during the 20th

4 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, E X E C U T I V E S U M M A RY

century reflect policies and planning that prioritized it over

alternative means of transportation. Although the auto-
mobile was originally seen as a technological advance that
would improve society as a whole, market conditions and
policies yielded a sprawling development pattern with nega-
tive environmental, social, and economic impacts—issues
that planners still wrestle with every day.
Today, AVs are poised to disrupt the built environment
and planning practices in ways that may be hard to imag-
ine and will be difficult to immediately determine. While
the private automobile yielded a 20th century dominated
by suburban expansion, this report makes the case that AV
technology has the potential to support and promote ur-
ban (re)development for the next century. With planners
leading the way, a sustainable AV future can be made pos-
sible through thoughtful visioning, quality planning, and
smart investment. Now is the time to begin embracing the
transformative power of autonomous vehicle technology to
change our communities for the better.

www.planning.org AMERICAN PLANNING ASSOCIATION 5

CHAPTER 1
INTRODUCTION
Just over a century ago, the mass production and rising affordability of the private automobile contributed to massive changes
in transportation networks, individual travel behaviors, and the built environment (Chapin, Stevens, and Crute 2017). The
private automobile provided greater mobility, allowing drivers and their passengers to escape dense, complicated, and pol-
luted urban centers. In the following decades, federal, state, and local governments supported this increased mobility through
massive investments in road networks, state highways, and ultimately a massive interstate system that allows for high-speed
travel over long distances.

This increased mobility brought about long-term chang- Stevens, and Crute 2016; Riggs forthcoming). This rapidly
es in the built environment as well. At the metropolitan level, advancing technology offers the promise of increased safety
reduced travel times and costs contributed to the suburban- for users and greater efficiency in systems operation. AVs also
ization of upper-class households in the 1920s, and then mid- allow riders to be productive and entertained during their
dle-class households starting in the 1940s. These moves took travel times, provide mobility to populations that are unable
shape for a number of reasons, from middle-class families to drive (children and the elderly), and will almost certainly
of returning GIs looking for larger homes to historic racial contribute to changes in vehicle ownership patterns. They
tension resulting in concentrations of minorities in and afflu- offer the potential to launch a wide range of new mobility
ent suburban flight from many large cities. In any case, these options that serve targeted populations ranging from urban
new suburbanites consumed far-flung single-family hous- communities to isolated lower-income populations.
ing at extremely affordable rates, contributing to America’s Alongside this potential, AVs (and artificial intelli-
sustained economic boom in the post-World War I and post- gence in general) offer possible challenges. They could lead
World War II eras. to changes in historical housing settlement patterns, caus-
As these wealthier households left the city, retail centers ing sprawl to spread farther into exurban areas. In a rapidly
followed. Whereas the early 1900s saw most retail outlets lo- aging society with growing social and economic disparities,
cated in downtowns and urban neighborhoods, by the 1960s they may exacerbate spatial and cultural separation. As e-
the suburban shopping mall and strip mall had come to dom- commerce and workplace automation continue, they could
inate the retail market. The rise of the automobile brought reshape how, where, and when community members live,
about changes at the corridor and site levels as well. Urban and work, play, and shop.
suburban arterials were designed to promote speed and safety Beyond these very important improvements in safety
for automobiles, and mobility concerns came to dominate and mobility, not to mention the possible quality-of-life
roadway design. At the site level, buildings were pushed back benefits for a variety of users, AVs have great potential to
to make room for automobile parking, and parking standards impact and alter the built environment in the coming de-
for the busiest parking day of the season came to dominate cades. While these impacts—positive or negative—are not
local codes. In a few decades, the American landscape was fully predictable, all indications are that the impact of AVs
largely redesigned to ensure that the private automobile could on roadway design, urban form, and site design may be of
travel at speed, over long distances, and with easy ingress and a magnitude similar to those that occurred during the rise
egress to the vast majority of land uses in the city. of the private automobile in the early 20th century. As de-
The central contention of this report is that autonomous tailed in the report, there is strong evidence that AVs could
vehicles (AVs) will cause the next great transformation in our enable narrower rights-of-way and travel lanes; influence the
transportation systems and the built environment (Chapin, location, form, and amount of parking; impact the mobility

www.planning.org AMERICAN PLANNING ASSOCIATION 7

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

of bicyclists and pedestrians; declutter urban environments need to harness the opportunities AV technology provides,
through reduced signalization and signage; and provide op- mitigate potential concerns, and ensure sustainable and peo-
portunities for redevelopment on excess parking lots and ple-oriented communities. Autonomous vehicles will cause
rights-of-way. the next great transformation, impacting not only transpor-
Nonetheless, it is important to recognize that AVs will tation systems, but the built environment of our cities. How
not solve all community problems. If increases in driving we respond will shape this impact.
continue and household car ownership remains as it is, more
cars driving more miles each year will only worsen conges-
tion issues. AVs bring the possibility of making biking and PLANNING, UNCERTAINTY, AND
walking more difficult, because AVs require more frequent THE REVOLUTION
stops and free-flow intersections may become ubiquitous.
Lastly, AVs may spell real problems for already challenged The adoption of and planning for AVs is surrounded by a
mass transit systems, and bus drivers, taxi drivers, and truck- large degree of uncertainty. All of the major car manufactur-
ers may see their work opportunities disappear. Planners will ers are actively working on AV technology and several have
need to anticipate and mitigate these new challenges to con- already tested vehicles on the roads. Technology companies
tinue building better communities. Consequently, planners such as Google and Apple also have vehicles, with others such

TABLE 1.1. POTENTIAL MANUFACTURERS, FUNDING, AND PLATFORM NOTES

Approximate Level 3 Level 4/5

Company Shared / Fleet Notes
Funding Availability Availability

BMW 2021 2025 Y Shared at first; partnership with Intel and Mobileye

Daimler /
TBD TBD Y Partnership with Uber; freight introduced first
Mercedes

Ford $1–2B 2021 TBD Y Potential partnership with Waymo or Lyft

Fiat-Chrysler 2021 TBD Potential partnership with Waymo or Lyft

GM $581M 2018 TBD Partnership with Lyft

Honda 2020 TBD Potential partnership with Waymo or Lyft

Hyundai $1.7B 2020 TBD Level 3 (highway capable) by 2020

Renault-Nissan 2020 TBD Partnership with Nissan

Claims Level 3 capability with autopilot and that Level 4 can

Tesla 2017 2020 N
be achieved without Lidar

Toyota $1B 2020 TBD TBD Being led by Toyota Research Institute

Waymo 2020 TBD May develop vehicles independently

Volvo $300M 2021 TBD Partnership with Uber; will self-insure for liability

Sources: Venture Beat.com, IDC, BMW

8 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

as Tesla claiming that their vehicles could be autonomous in expanded roadways and parking garages, but AVs may
(Crothers 2016; Harris 2015). change the demand for parking and spatial siting variables.
There are still questions about when vehicles will be de- City planners and leaders should begin asking questions and
ployed and ready for purchase. While much of these are spec- preparing for this potential future. By looking at different
ulation, as shown in Table 1.1 (p. 8), synthesized data from AV-future scenarios, planners can develop plans that are flex-
numerous sources suggest the most basic form of AVs will ible and adaptable.
be widely available in three to five years. This basic level of
autonomous driving is called Level 3 technology, which is the
equivalent of “lane assist” or systems that correct a vehicle’s ABOUT THIS REPORT
course when it drifts out of its lane. (More detail about the
different “levels of autonomy” is provided in Chapter 2.) The primary goals of this PAS Report are to (1) provide plan-
While this report provides more information about these ners and policy makers with the foundational knowledge
levels of autonomy, ultimately, the availability of AVs and necessary to anticipate potential impacts of AVs on commu-
when they will become fully autonomous is subordinate to nities and (2) support and spur development of policy solu-
how they will be owned and used in the future. This relates to tions and infrastructure investments that ensure attractive,
three concepts that will be discussed as a part of this report— people-friendly, equitable, and safe urban environments.
what researchers have referred to as the “three revolutions In this report we attempt to envision the impact of AVs
in urban transportation” (3Rs) (Fulton, Mason, and Meroux on communities as a starting point for planning agencies to
2017). Planners have speculated that three key revolutionary begin preparing and planning for the emergence of AV tech-
aspects of AVs (that they will be autonomous, electric, and nology. This report focuses on issues and policy interven-
shared) will impact cities. Yet, the future is uncertain and a tions for planners to consider so that they can develop more
3R scenario is highly dependent on the decisions automak- thoughtful, robust, and adaptable plans to prepare for the
ers (sometimes called original equipment manufacturers, or adoption of AVs. Planners can begin rethinking things such
OEMs) and land-use planners make now. as parking, street design and engineering, streetscape and ur-
In light of this, this report argues that the current un- ban design, asset investment, municipal finance, transit and
certainty is not an excuse for inaction. There has been very bikes, and land use, among others.
little policy development addressing the potential benefits or The emergence of AVs is almost upon us, and how the AV
limitations of the AVs of the future. Moreover, the policy that revolution takes hold largely depends on the actions planners
has been developed is highly speculative. and policy makers take now. Consequently, planners have an
The planning and infrastructure projects that planners important opportunity to shape sustainable, resilient urban
provide guidance on shape the form of cities every day, and forms where AVs contribute to a successful multimodal sys-
these recommendations and insights have long-term impli- tem. The structure and content of the report is outlined below.
cations. Agencies like the World Economic Forum already Chapter 2 provides a primer of baseline information on
speculate that the technology is developing faster than ex- the state of AV technology today. For planners to make in-
pected and that cities will likely not be prepared for self-driv- formed policy decisions, they need to understand the capabili-
ing vehicles (Abrams 2016). Work by Guerra (2015a, 2015b) ties and limitations of the technology. To this end, this chap-
found that in 2015, only two of the 25 largest metropolitan ter describes what AV technology is and is not capable of. It
areas mentioned autonomous or connected vehicles in their then outlines the implications of several other advancements
planning documents. This work cites (1) the uncertainty of in transportation technology, such as connected vehicles and
the impact of AVs and (2) the disconnect between present in- advanced traffic management systems that could complement
vestments and future technology as two of the primary rea- AVs and amplify their benefits. The convergence of autono-
sons why governments are not planning for the AV systems mous technology with car- and ride-sharing trends could shift
of the future. See the sidebar on pp. 10–11 for examples of the predominant model of automobile use from private own-
planning policy language addressing AVs that does exist. ership to shared mobility, particularly in urban areas. This
Planners and policy makers should be anticipating a chapter highlights how the use of shared AVs could further
changing and uncertain future by practicing scenario plan- their impacts on travel behavior and the built environment.
ning and providing incremental guidance. For example, Finally, Chapter 2 draws from professional and academic pro-
many communities are making significant investments jections to provide an expected timeline for AV adoption.

www.planning.org AMERICAN PLANNING ASSOCIATION 9

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

AV POLICY LANGUAGE FROM EXISTING PLANNING DOCUMENTS

Austin, TX, Smart Mobility Roadmap: 13. Create a regional New Mobility Work- 2. Collaborate regionally to promote in-
Austin’s Approach to Shared, Elec- force Training task force for new job teroperability.
tric, and Autonomous Vehicle Tech- training and educational opportuni- 3. Launch a taskforce on data monetiza-
nologies (2017) ties for those with legacy occupations tion strategies.
www.austintexas.gov/smartmobility 4. Advocate for new approaches to fi-
roadmap Boston Transportation Department, nancing infrastructure projects.
Go Boston 2030 Vision and Action TODAY (0–2 years)
Autonomous Vehicles Recommended Plan (2017) 1. Develop a business plan for a city AV
Actions: w w w. b o s to n . g o v/d e p a r t m e nt s / fleet.*
1. Engage citizens, businesses and transportation/go-boston-2030 2. Create a dedicated staff position fo-
visitors on how this technology can cused on connected and automated
meet their needs and address com- Goal: Flexibility to accommodate disrup- vehicle tech.
munity issues tive mobility technologies 3. Implement blind spot detection sys-
2. Hire an Executive Level Officer of EV/ The arrival and adoption of new tems for public transit vehicles.*
AV Transportation technology—such as autonomous 4. Expand LADOT connected bus tech-
3. Develop a Master Plan roadmap for cars, electric tricycles, and self-driving nologies fleet-wide.
emerging electric–connected and buses—is imminent. Boston will ac- 5. Invest in lane markings that enhance
autonomous vehicle (E-CAV) tech- commodate these and other emerging effectiveness of lane departure warn-
nologies vehicle types by creating infrastructure ing and prevention systems.
4. Create an interdisciplinary AV Work networks that can be easily repurposed. TOMORROW (3–5 years)
Group Car and curbside lanes on major corri- 1. Create better access to ATSAC data
5. Create an infrastructure task force to dors like Columbia Road or in dense ar- and enhance transparency of net-
examine electric, technology and eas such as the Theater District will offer work prioritization for planning.
land use infrastructure requirements parking at some times and bus or bike 2. Develop an AV road network along
6. Test Dedicated Short Range Commu- lanes at others and serve as designated transit and enhanced vehicle net-
nication (DSRC) technology for ve- pick-up and drop-off locations for pas- works.
hicle to infrastructure (V2I) reciprocal sengers and parcels. Traffic signals will 3. Launch a Data as a Service program
safety messages adapt automatically, relying on sensors to provide real-time infrastructure
7. Test 5G technology for vehicle to and algorithms to optimize the move- data to connected vehicles.
infrastructure (V2I) reciprocal safety ment of people. New buses will be com- FUTURE (6+ years)
messages; compare to DSRC patible with older fleet vehicles while • Convert the public transit vehicle
8. Increase public awareness of electric leveraging emerging technology. fleet to fully automated.
autonomous (E-AV) shuttles in vari- ———
ous Austin locations through EV/AV Los Angeles Department of Trans- * Action already planned or underway.
pilots portation, Urban Mobility in a Digital
9. Increase public awareness of last mile Age (2016) Portland, OR, Draft Connected and
E-AV delivery robots www.urbanmobilityla.com/download Autonomous Vehicles Policy (2017)
10. Establish an EV/AV Commercialization www.portlandoregon.gov/transporta-
Opportunities/ Economic Develop- Transportation Technology Strategy 5: tion/article/643814
ment Work Group Prepare for an automated future
11. Create Shared/EV/AV focused team Policy 9.xx Connected and Autono-
12. Increase public awareness of electric Policy Recommendations mous Vehicles. Ensure that connected
and autonomous vehicle benefits 1. Call for mobility innovation in California. and autonomous vehicles advance

10 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

Portland’s Comprehensive Plan mul- Policy EA6: Acknowledge and miti- privacy, open, anonymized data can im-
tiple transportation goals and policies, gate the labor implications of automat- prove the City’s decisionmaking and
including vision zero, climate pollution ed mobility, particularly in the for-hire, help to develop more informed policies
reduction and cleaner air, equity, physi- freight, and public transit industries, and plans.
cal activity, economic opportunity, great among others. » Introduce polices that can influ-
places, cost effectiveness, mode share, Policy EA7: Conduct a publicly-vis- ence how driverless vehicles can affect
and reducing vehicle mile traveled. ible community consultation and out- VMT, urban sprawl, and/or parking re-
reach process to understand concerns, quirements. Examples include tolls for
Seattle Department of Transporta- needs, and opportunities related to the single-occupancy vehicles, new HOV/
tion, New Mobility Playbook, Appen- impending automated mobility para- HOT lanes, create and enforce urban
dix C: Preliminary Automated Mobil- digm. growth boundaries, reduce (or even sub-
ity Policy Framework (2017) Policy EA8: Establish a City-owned sidize) costs and parking fees for shared
https://newmobilityseattle.info transportation network company digi- ride services, and explore parking re-
tal platform to incubate smaller shared quirements in zoning laws and encour-
EQUITY AND ACCESSIBILITY automated vehicle fleet businesses, miti- age more pick-up/drop-off locations at
The following policies ensure that auto- gating the risk of mobility monopolies in developments.
mated mobility and other future trans- Seattle
portation innovations are designed with [Other policies address regula- San Jose, CA, Smart City Vision
a racial and social justice lens, accommo- tion and parameters, infrastructure and www.sanjoseca.gov/index
dating the wide cross section of Seattle- street design, pilots and partnerships, .aspx?NID=5289
ite’s abilities and backgrounds. mobility economics, and land use and
Policy EA1: Ensure the benefits of building design.] Demonstration City: Reimagine the City
automated mobility are equitably dis- as a laboratory and platform for the most
tributed across all segments of the com- San Antonio, TX, SATomorrow Multi- impactful, transformative technologies
munity and that the negative impacts of modal Transportation Plan (2016) that will shape how we live and work in
automated mobility are not dispropor- www.satransportationplan.com the future.
tionately borne on traditionally margin- Fully develop the city’s transporta-
alized communities. The City of San Antonio should consider tion innovation zone to test new prod-
Policy EA2: Ensure shared auto- the following planning and policy activi- ucts and services, such as autonomous
mated vehicle fleets consider the safety ties to manage the impact of CV/AV on vehicles, that will dramatically shape
needs of vulnerable populations and the city: transportation in the future and mitigate
loading needs of seniors, families with » Update the City’s travel demand traffic congestion.
children, and individuals with mobility model. The City’s travel demand mod- Build an “Internet of Things” plat-
impairments. els should ideally reflect updated infor- form employing transit vehicles and in-
Policy EA3: Establish equitable per- mation regarding who is traveling (e.g., frastructure by using smart sensor tech-
formance standards and penalty struc- elderly and disabled may travel more nologies to improve safety, mobility, and
tures for shared automated vehicle fleet due to AVs), where people are living and optimize our transit system.
wait time and declined rides as a way to working, how many trips they are taking, Create pathways for start-ups and
eliminate discriminatory practices. people’s value of time while traveling, innovators to easily access opportuni-
Policy EA4: Require a percentage of what level of shared rides are occurring, ties to pilot and test new products and
shared automated vehicle fleet vehicles and the vehicle ownership model. It services with the City, such as by hosting
to be ADA-compliant to meet the needs should also capture any changes asso- “demo days” to highlight the most inno-
of people with disabilities. ciated with freight delivery. This update vative “smart city” companies in Silicon
Policy EA5: Identify and require needs to be on the City’s horizon as the Valley, and sponsoring public competi-
shared automated vehicle fleets to serve industry matures its approach to fore- tions to encourage crowdsourcing of
markets that are underserved by transit casting this new future. innovative solutions to civic challenges.
and focus on connecting people to high » Encourage open data sharing.
quality transit spines. While it is important to preserve people’s

www.planning.org AMERICAN PLANNING ASSOCIATION 11

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

Chapter 3 outlines the major opportunities and chal- opted, this chapter also emphasizes the need for nimble plan-
lenges likely to emerge as AV technology becomes ubiquitous ning processes and policies that proactively accommodate the
in communities around the country. AVs have the potential technology’s rapidly evolving capabilities. Key considerations
to improve the safety and efficiency of transportation sys- to incorporate into infrastructure investments and redevel-
tems, reduce vehicle emissions, and improve the mobility of opment decisions moving forward are also highlighted. Fi-
transportation-disadvantaged populations. Unfortunately, nally, Chapter 6 wraps up the report with a final call to action.
AVs will also bring challenges that threaten to negate their
potential benefits. In particular, AVs may reinforce auto-
oriented sprawl, which could increase vehicle miles traveled
(VMT) and congestion. Without careful planning, AVs could
also compromise bicycle and pedestrian mobility. Further,
AVs will have important ramifications for several key urban
issues, such as public transit, public health, and social equity.
Since private companies are investing so many resources in
developing the technology itself, these secondary effects of
AVs will be the main concern for planners to ensure the tech-
nology does not have adverse ramifications for placemaking
or quality of life. This chapter summarizes each of these is-
sues to identify how they will shape AVs’ impact on the built
environment and appropriate policy responses to capitalize
on the important opportunities the technology provides.
Building on the findings of the first three chapters, Chap-
ter 4 makes the case that AV technology will catalyze the next
great transformation in the built environment. This section
draws heavily upon a Florida Department of Transportation-
funded study completed at Florida State University (Chapin,
Stevens, and Crute 2016) to identify six major areas where
AVs may impact the built environment: (1) new designs of
rights-of-way, (2) changes to access management practices,
(3) reconsideration of the form and function of signage and
signalization, (4) new models for pedestrian and bicycle net-
works, (5) reductions in demand and changes to the location
of parking, and (6) new redevelopment opportunities in ur-
ban and suburban locales. Each of these is explored in detail
to develop a potential vision of the future in an AV world.
Chapter 5 then provides guidance on how planners
should prepare for and respond to these far-reaching chang-
es. This will provide a place for communities to start to ad-
dress the planning opportunities and challenges identified
in Chapter 3 and the ways AVs will shape the built environ-
ment described in Chapter 4. Throughout this discussion, the
chapter highlights the need for proactive planning efforts to
ensure that future development patterns and urban form are
shaped by sound planning principles rather than by the tech-
nology. In other words, cities should be designed for people
and not for technology to be attractive, people-friendly, eq-
uitable, and safe urban environments. Given the uncertainty
surrounding how and when AVs will be developed and ad-

12 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 1

www.planning.org AMERICAN PLANNING ASSOCIATION 13

CHAPTER 2
AUTONOMOUS
VEHICLES 101
Autonomous vehicle technology is a rapidly developing technology that promises to revolutionize the form and function of
our urban spaces. To understand how it will do that, it is important to become familiar with what the technology is and how it
works. AV technology is often presented as an easy solution to many of our planning problems. Understanding what the tech-
nology is and is not capable of will enable planners to more effectively use it as a tool to improve their communities without
expecting it to solve safety and congestion issues or neglecting sound planning practice.

AV technology is advancing so rapidly that it is vital for is published. Therefore it is important that this chapter is
planners to stay up to date on the latest in technological ad- viewed not as the definitive guide to AVs but as a first step in
vancements. This is especially important because, as will be a continuous learning process.
described in more detail later, the exact form and capability
of the technology will ultimately be a major determinant of
its impact upon the transportation system and the built en- THE TECHNOLOGY: WHAT IS IT AND
vironment. For example, the size of AVs may determine lane HOW DOES IT WORK?
width and other roadway design features. Consequently, this
section will provide an overview of the current state of the AVs have captured the public’s imagination and have been the
technology, but given the speed at which the technology is main focus of the media’s discussion of intelligent transpor-
advancing, this may be out of date by the time this report tation systems. However, recent advancements in technology

Figure 2.1. Diagram

of automated vehicle
GPS (global positioning system) technologies (adapted
combined with readings from Lidar (light detection and ranging)
tachometers, altimeters and monitor the vehicle’s surroundings (road, from BCG 2015)
gyroscopes to provide the most vehicles, pedestrians, etc.)
accurate positioning

Video cameras
monitor the vehicle’s surroundings
Ultrasonic sensors to (road, vehicles, pedestrians, etc.) and
measure the position of objects read traffic lights
very close to the vehicle

Odometry sensors to
complement and improve
GPS information

Central computer analyzes all sensor input, Radar sensors

applies rules of the road and operates the steering, monitor the vehicle’s surroundings
accelerator and brakes (road, vehicles, pedestrians, etc.)

www.planning.org AMERICAN PLANNING ASSOCIATION 15

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

have included much more than just AVs. The development

DEFINITIONS AND ACRONYMS of advanced driver assistance systems and connected vehicle
technology has quietly provided growing opportunities to
Autonomous Vehicle (AV)— A vehicle that improve the safety and efficiency of the transportation sys-
is capable of driving itself without hu- tem. Ultimately, the greatest benefits will come from having
man intervention all these technologies and systems working together in har-
mony, but while fully autonomous vehicles are being tested
Advanced Driver Assistance Systems and piloted, these additional technologies may provide excit-
(ADAS)—A range of vehicle technolo- ing opportunities, particularly in the near term. Consequent-
gies that enhance driver safety by tak- ly, it is important for planners to be aware of the full range of
ing temporary control of one or more available technologies to make informed decisions as to what
driving functions (speed, lane position, will provide the most benefit for their communities. The side-
braking, etc.) bar on this page offers a list of definitions and acronyms as a
resource for planners on AV terminology.
Connected Vehicle (CV)—Features that
enable vehicles to communicate with
Automated Vehicle Technology
other vehicles, the infrastructure, or pe-
Automated vehicle technology is an umbrella term that in-
destrians
cludes a wide variety of features and technologies that enable
Vehicle to Vehicle (V2V)—CV technol- vehicles to take control of some or all of the major driving
ogy that enables vehicles to commu- functions normally completed by the driver (Figure 2.1, p.
nicate with other vehicles 15). This includes fully autonomous vehicles that no longer
require a human driver to operate them, as well as a range
Vehicle to Infrastructure (V2I)—CV
of advanced driver assistance systems (ADAS) that enhance
technology that enables vehicles to
driver safety by taking temporary control of one or more
communicate with the smart infra-
driving functions (speed, lane position, braking, etc.).
structure
A fully autonomous vehicle no longer requires a human
Vehicle to Pedestrian (V2P)—CV tech- operator to drive. Instead, the vehicle navigates streets safely
nology that enables vehicles to com- and efficiently through a complex mix of software and hard-
municate with nearby pedestrians ware that combines remote sensing, recognition algorithms,
network analysis, and “experience” drawn from millions
Vehicle to Everything (V2X)—CV tech-
of hours of driving that is shared across AVs. The vehicle’s
nology that enables all vehicles and
combination of sensors, cameras, light detection and ranging
infrastructure to be interconnected
(Lidar or light radar), high-definition maps, and advanced
Intelligent Transportation System (ITS)— software create a digital picture of its surroundings and make
Advanced technologies that improve
the safety and efficiency of the transpor-
tation system by collecting, analyzing,
and communicating information in real
time
Connected Autonomous Vehicle (CAV)—A
vehicle that has both connected and au-
tonomous capabilities
Shared Autonomous Vehicle (SAV)—An
autonomous vehicle that is available on
a short-term, “as-needed” basis

Figure 2.2. Example of a Lidar cloud (source: Waymo)

16 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

intelligent driving decisions on routing and maneuvering vehicle automation (Figure 2.3, p. 18). These range from “0,”
without any input from an operator or information broadcast where the driver is in complete control of all driving tasks at
by infrastructure or other vehicles. all times, to “5,” where the vehicle is designed to perform all
More specifically, just as radar does with radio waves, Li- driving tasks without an operator (SAE International 2016).
dar shoots pulses of light and measures how long it takes for the With Level 1 automation, the driver remains in control
light to return to the sensor to assess how far away an object is. of the vehicle, but the technology can assist the driver by con-
As seen in Figure 2.2 (p. 16), placing an array of rotating lasers trolling one of the vehicle’s functions, either its speed or lane
on top of an AV provides a continual 360-degree “point cloud” position. Level 2 takes this a step further by allowing the ve-
or picture of the vehicle’s surroundings. The vehicle’s central hicle to control two driving functions at the same time. A ve-
computer can then be programmed to recognize specific Lidar hicle with Level 3 automation can take full control of the ve-
returns as another car, a pedestrian, or even a stop sign. hicle for certain parts of a trip, but the driver must be ready to
Lidar systems are typically supplemented by cameras take back control of the vehicle when the vehicle prompts her.
and other sensors to provide redundant detection systems The vehicle takes full control of all major driving functions in
that will not fail to detect objects that Lidar could miss, par- Level 4. Level 4 vehicles can even drive themselves for the en-
ticularly in the area immediately surrounding the vehicle. tire trip, but they are only able to do so under specific condi-
More sophisticated systems add another layer to this by as- tions. Finally, Level 5 automation refers to fully autonomous
sessing how surrounding vehicles and pedestrians are mov- vehicles that can operate without an operator in all conditions
ing and predicting where they will go next. In the case of a and without the capability for a human to retake control.
pedestrian crossing the street, the vehicle can predict the Automated driving features that aid the driving process
pedestrian’s movements and begin slowing down before the but do not fully control the vehicle (Levels 0, 1, and 2) are gen-
pedestrian enters the street instead of waiting until the pedes- erally referred to as ADAS. Even though fully autonomous
trian is directly in the vehicle’s path. vehicles have received most of the attention and are the focus
Unfortunately, whether an AV uses Lidar or cameras or of this report, ADAS can significantly improve driver safety,
both, it is very difficult for these systems to work properly in thereby improving user mobility. For example, one of the
inclement weather conditions and poor visibility. Rain and most common crash scenarios among aging drivers is mis-
snow can refract the laser returns and cameras struggle to judging oncoming traffic while making a left turn. Simula-
identify objects accurately through precipitation, function- tor studies have shown that even a simple Level 0 automation
ally blinding the AV. However, using both technologies in feature that informs drivers when they have enough space to
tandem could overcome this problem as the technology con- turn left could significantly improve the safety of aging driv-
tinues to advance. ers, thereby enabling them to continue driving and maintain
Most of the attention on AVs is centered around fully their personal mobility later into life (Davidse 2006).
autonomous vehicles because many of the technology’s most Most of the first applications of AV technology will be
significant effects on the transportation system and the built increasingly sophisticated ADAS. Even Tesla’s Autopilot fea-
environment will only be viable when fully autonomous ve- ture, introduced in 2015, would be classified as Level 2 auto-
hicles are adopted. However, AV technology includes a range mation, as it only controls the vehicle’s speed and lane posi-
of levels of automation. It is important for planners to be fa- tion and requires the driver to be “in control of the car” at
miliar with the full array of AV technology, because many all times (Tesla 2015). Consequently, it is vital for planners
semiautonomous features and applications are already avail- to be aware of the development and use of these features to
able today and will likely play a major role in the transition to take advantage of the benefits they can provide and to effec-
a fully autonomous world. tively manage the transition from human-driven vehicles to
In addition to fully autonomous vehicles, there is a wide vehicles equipped with ADAS to fully autonomous vehicles.
range of automated technologies that can operate as stand-
alone features. These range in sophistication and complex- Connected Vehicle Technology
ity from cruise control to autopilot. To classify these ever- Connected vehicle (CV) technology includes the vehicles and
evolving technologies, the National Highway Traffic Safety infrastructure that enable vehicles to communicate with other
Administration (NHTSA) and the Society of Automotive vehicles, infrastructure, or pedestrians to make better driving
Engineers (SAE) International developed a classification decisions. CV technology relies on information gathered by
system that divides automated technologies into six levels of vehicles and the transportation infrastructure about real-time

www.planning.org AMERICAN PLANNING ASSOCIATION 17

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

Figure 2.3. The levels

of vehicle automation
Five Levels of Vehicle Autonomy
(SAE and NHTSA)

Level 0 Level 1 Level 2 Level 3 Level 4 Level 5

No automation: Driver Assistance: Occasional Limited Full self-driving Full self-driving
the driver is in the vehicle can self-driving: self-driving: under certain under all
complete control assist the driver or the vehicle can take the vehicle is in full conditions: conditions:
of the vehicle at take control of control of both the control in some the vehicle is in the vehicle can
all times. either the vehicle's speed and situations, full control for the operate without
vehicle's speed, lane position in monitors the road entire trip in these a human driver
through cruise some situations, for and traffic, and conditions, such or occupants.
control, or its lane example on will inform the as urban
position, through limited-access driver when he or ride-sharing.
lane guidance. freeways. she must take
control. Source: SAE & NHTSA

operations of the transportation network. Based on a specific they begin slowing down or turning. This would enable AVs
vehicle’s location, information is broadcast to the vehicle so to travel in even safer harmony.
the driver is able to make informed decisions regarding rout- Like ADAS, applications of CV technology will be im-
ing and maneuvering. Yet, by itself, this technology does not plemented well before fully autonomous vehicles are adopt-
impact safety-critical functions of the vehicle and the driver ed. In 2016, the U.S. Department of Transportation awarded
must remain in full control of the vehicle at all times. New York City, Tampa, and Wyoming more than $45 million
Simple examples of CV technology include transmit- in collective funding to “design, build, and test” operational
ting information typically given on street signs to a heads- CV systems through the CV Pilot Development Program
up display in the vehicle. For instance, a sensor embedded (U.S. DOT 2016). As part of this program, Tampa will have
in the roadway could tell the vehicle what the speed limit is 10 buses, 10 streetcars, and 1,600 personal vehicles equipped
at all times or it could provide a warning whenever the ve- with CV technology on the road by 2018 (Tampa Hillsbor-
hicle begins traveling the wrong way down the road. More ough Expressway Authority 2017). In addition, companies
sophisticated examples could include an ambulance warning such as Peloton and Daimler are already piloting connected
other vehicles to move out of the way or platooning, in which semi-truck applications on highways across the country that
two or more vehicles “link” and travel together like a train. promise to significantly improve the fuel efficiency of the
The driver remains responsible for using this information to trucking industry.
operate the vehicle, but the information provided helps the CV technology is generally divided into three major
driver to make safer and better-informed driving decisions. types: Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure
AV and CV technology could each provide positive safety (V2I), and Vehicle-to-Pedestrian (V2P) communication.
and efficiency benefits to the transportation system on their These technologies are not mutually exclusive, meaning that
own, but it is commonly accepted that the most significant a single vehicle can be equipped with more than one type of
benefits will only be achieved by vehicles that are both au- CV. Vehicles equipped with all three are considered to have
tonomous and connected. For example, a fully autonomous Vehicle-to-Everything (V2X) capabilities, as these vehicles
vehicle can safely navigate a traffic jam, but a connected AV would be able to communicate with anything that may affect
could avoid the traffic jam altogether by finding the fastest their operation in real time. In many cases, several redundant
alternative route in real time. In addition, CVs could inform systems will likely need to be in place to ensure uninterrupted
other vehicles they intend to brake or change lanes before communication in the event of a system failure.

18 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

Vehicle-to-Vehicle Communication and CV technology is advancing so rapidly that detection

V2V technology includes systems that allow vehicles to com- accuracy is expected to improve dramatically, but V2P com-
municate their location, speed, heading, and other informa- munication that informs the vehicles of a bicyclist’s location
tion to other vehicles on the road. V2V communication can could add a vital layer of redundancy to a vehicle’s detection
add an important redundancy to AV sensor data, thereby system that may enable vehicles to safely navigate a dynamic
ensuring the vehicle detects surrounding vehicles even if the urban environment.
Lidar sensors fail to identify this information. More impor-
tantly, it can enable AVs to travel in harmony with each other, Other Technological Advances
further improving the efficiency of an automated transporta- That Will Affect AVs
tion system. By enabling two or more vehicles to link together In addition to AV and CV technology, there are several other
in a platoon, CV technology can eliminate the time it takes to transportation technology advancements that could comple-
detect a change in another vehicle’s movement and allow ve- ment the emergence of AVs and magnify the benefits AVs
hicles to safely travel even closer together, further improving promise to provide.
fuel efficiency by reducing wind resistance.
Electric Vehicles
Vehicle-to-Infrastructure Communication Many proponents of AVs point to automated vehicle technol-
V2I communication enables vehicles to exchange informa- ogy, electric vehicle (EV) technology, and on-demand ride
tion with roadway infrastructure and traffic management sharing as three converging trends that promise to undo
systems. The infrastructure can provide vehicles with data many of the problems created by our current transportation
and information to make safer and more efficient driving de- system. By providing a more efficient transportation system
cisions by informing the vehicle of traffic signs and signals, that utilizes cleaner energy sources, on-demand electric AVs
road conditions, traffic incidents, and optimal routes. Simi- could drastically reduce energy emissions. In fact, many of
larly, the vehicle can communicate with the infrastructure the environmental and efficiency benefits AVs could provide
to facilitate more efficient traffic operations. For instance, will only be possible if AVs are powered by electricity. Oth-
vehicles can inform the infrastructure of the location of traf- erwise, the transportation system will remain completely
fic congestion or hazardous roadway conditions such as icy dependent on combustion engines and AVs at best will be
roads or standing water. A traffic management system could a temporary fix for the global carbon emissions crisis and
warn other vehicles to avoid these areas or find faster routes. at worst will exacerbate the problem by increasing vehicle
miles traveled (VMT).
Vehicle-to-Pedestrian Communication Many have speculated that because AVs will require a
CVs could also communicate with bicyclists and pedestrians. significantly more extensive electrical system to power the
The majority of V2P systems provide the vehicle operator sensors and computers necessary to drive autonomously,
with visual and audible warnings that a pedestrian may move it will be easier and more efficient to engineer AVs that are
into the vehicle’s path to ensure the driver or the AV detects electric-powered rather than combustion-powered (Gardner
and reacts to pedestrians. Some of these systems also provide 2016). The fact that many AV prototypes on the road today,
bicyclists and pedestrians with warnings of oncoming traffic including models being used by Waymo, Uber, and Lyft, are
via their cell phones. V2P communication could be especial- hybrid vehicles may provide an early sign of how AV technol-
ly important because detecting bicyclists and pedestrians is ogy may complement the increasing development and use of
one of the most difficult challenges for AV technology (Fairly EVs. Similarly, recent announcements from General Motors,
2017). Bicyclists and pedestrians are “small, unpredictable, Volvo, and other major auto manufacturers of their inten-
and hard for computers to see” (Sorrel 2017). tions to work toward an all-electric future (Davies 2017) may
While pedestrian detection data for companies like provide a positive indication of the potential for a fully elec-
Waymo (formerly Google’s self-driving project) remains pro- tric and fully autonomous future. Many of these announce-
prietary, a Deep3DBox three-dimensional object detection ments have been driven by the major innovations in battery
algorithm recently developed by researchers from George technology that have significantly improved the cost effec-
Mason University, in conjunction with the robotic taxi com- tiveness of electric vehicles. The cost of EV battery packs fell
pany Zoox, was only able to successfully identify bicycles 74 by 80 percent between 2010 and 2016 (McKinsey & Company
percent of the time (Fairly 2017; Mousavian et al. 2017). AV 2017). While the per-mile cost of EVs will need to continue to

www.planning.org AMERICAN PLANNING ASSOCIATION 19

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

drop for them to be a viable option for mass adoption, these ity model. While AVs will be a transformative technology
trends are projected to continue and AV production is ex- regardless of the ownership model, using AVs in a shared
pected to largely comprise EVs (Collie et al. 2017). system would amplify their impact on travel behavior and
the built environment.
Advanced Traffic Management Systems
Advanced traffic management systems (ATMS) utilize intelli- The Rise of the Sharing Economy
gent infrastructure and real-time traffic data to improve traf- and On-Demand Mobility
fic flow and vehicle safety. Different versions of ATMS have The digital age and the connectivity it provides opened the
been in use for decades; however, AV and CV technology door for the emergence of the sharing economy. Often de-
will create opportunities to significantly increase the utility fined as the “peer-to-peer based activity of obtaining, giving,
of ATMS. Today, most ATMS use traffic data gathered from or sharing access to good and services” (Hamari, Sjöklint,
sensors and cameras embedded in the roadway infrastruc- and Ukkonen 2015), the sharing economy has exploded into
ture to make adjustments to speed limits, traffic light tim- a $18.6 billion phenomenon (Juniper Research 2017). One of
ing, and ramp metering to improve vehicle flow and mitigate the major drivers of the sharing economy has been the emer-
traffic congestion. CV technology could revolutionize ATMS gence of the shared mobility industry. While shared mobil-
by enabling them to be informed by data inputs from every ity services such as ZipCar, Uber, and Lyft make up a small
vehicle on the road. fraction of total VMT today, the convergence of recent demo-
Yet even more transformative is the potential for ATMS graphic trends and the emergence of AV and EV technologies
to communicate with or even control the movements of ve- may help shared mobility become an increasingly popular
hicles across the transportation system. All data and infor- and cost-effective option.
mation collected by AVs and CVs theoretically could be fed Recent trends of declining car ownership rates among
into a centralized ATMS that could provide each vehicle with younger generations who prefer a multimodal urban lifestyle
optimal routing information based on real-time traffic condi- have been well documented. However, cost and convenience
tions. The ATMS could safely reroute traffic away from traffic are the primary drivers of mode choice, and until shared
jams and safety hazards and could ensure AVs are aware of mobility options become cheaper than owning a vehicle, the
road work, detours, and new roadway infrastructure. In this majority of the population is unlikely to shift to on-demand
way, the combination of AV, CV, and ATMS technology could mobility options for daily travel. According to a study by
maximize the efficiency of a city’s roadway infrastructure in Morgan Stanley (2016), shared mobility options today cost al-
real time. While ATMS with this level of complexity have yet most twice as much per mile as vehicle ownership on average.
to be developed and may raise important privacy issues, it is Yet, AV technology may significantly alter the cost structure
clear that AV and CV technology will create exciting oppor- of vehicular transportation. First, the additional technology
tunities for ATMS to improve the safety and efficiency of the may significantly increase the price of purchasing a vehicle,
transportation system. making vehicle ownership less viable for individuals. Until
the price of an AV comes down, the high costs of purchasing
a vehicle may primarily be feasible for mobility companies
AV OWNERSHIP AND THE RISE OF SHARED AVS like Uber and Lyft, which can utilize and monetize the ve-
hicle 24 hours per day. Secondly, AVs may significantly re-
As the adoption of AV technology transforms the nature duce the cost of on-demand mobility service. By removing
of transportation safety and travel demand, changes to the the driver, and therefore the labor costs, AVs may signifi-
predominant automobile ownership model are likely to fol- cantly reduce shared mobility providers’ operational costs.
low. Up until recently, with the exception of taxis, almost This may decrease the price of on-demand mobility. Conse-
everyone owned the car they drove. However, the rise of ride- quently, numerous studies have projected that the per-mile
sharing services like Lyft and Uber have begun to challenge cost of shared autonomous mobility will become significantly
the traditional vehicle ownership model by offering a shared, cheaper than the cost of driving a personal vehicle (Burns,
on-demand mobility system. Jordan, and Scarborough 2013; Johnson and Walker 2017;
The convergence of ride sharing with autonomous driv- Bosch et al. 2017).
ing has the potential to shift the predominant automobile Such a shift in transportation costs could create a dra-
ownership model from private ownership to a shared mobil- matic shift in vehicle ownership rates, particularly in urban-

20 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

ized areas where on-demand mobility is more feasible. Con- notably, it would radically decrease the demand for park-
sequently, the Brookings Institution has estimated that the ing. Even if private vehicle ownership remains the norm, re-
sharing economy will grow to become a $335 billion market searchers from the University of Michigan found that private-
segment by 2025 and that this growth will be driven in large ly owned AVs could reduce vehicle ownership by 43 percent
part by the shared mobility industry (Yaraghi and Ravi 2017). (from 2.1 to 1.2 vehicles per household) (Schoettle and Sivak
Companies like Uber and Lyft are racing to capitalize 2015). Thus, regardless of whether AVs facilitate the rise of a
on this massive opportunity by developing their own AV shared ownership system, AVs are expected to reduce the size
technology. With ongoing autonomous testing in several of the vehicle fleet. Yet the impact to the vehicle fleet would be
U.S. cities, these ride-hailing companies are near the fore- significantly greater if they operated within a shared system.
front of AV development and implementation. Major auto However, reducing the size of the vehicle fleet does not
manufacturers have also recognized that their business necessarily mean that shared AVs would lead to a reduction
model may need to shift from private to shared ownership. in VMT or traffic congestion; each vehicle would simply trav-
In the last couple of years, multiple manufacturers includ- el significantly more miles per year. In fact, the introduction
ing GM, Toyota, and Volkswagen have invested hundreds of of “empty vehicle miles” while the vehicle is traveling from
millions of dollars into ride-hailing services (General Mo- one passenger to another will likely increase VMT. It is dif-
tors 2016; Bomey and Woodyard 2016). The CEO of Ford ficult to predict exactly how much empty vehicle miles would
Motor Company (2015) even announced a future vision to contribute to VMT because it will depend on the efficiency of
be “both a product and a mobility company.” the shared system as a whole. Total VMT could even decrease
The shift toward shared AVs does not mean that private if enough passengers were willing to share rides in addition
car ownership will completely disappear. There likely will be to sharing the vehicle. Ride sharing is generally considered to
a mix of shared and privately owned vehicles on the road, and be less popular than car sharing, but the relative success of
private ownership will likely remain the dominant model in UberPool may indicate that ride sharing may become more
rural areas. Yet shared vehicles are expected to make up a sig- popular when it is cheaper than car sharing, as UberPool
nificantly larger percentage of the vehicle fleet than they do made up 20 percent of Uber rides in 2016 (Lunden 2016).
today. As such, planners will need to be prepared for a grad-
ual shift toward shared ownership and the effects this may
have on congestion, parking, and urban design. TIMELINE FOR AV ADOPTION

Implications of a Shared AV System It is vital for planners to understand when fully autonomous
Moving toward a shared mobility model will have a host of vehicles will become available to the public and how long it
ramifications for travel behavior, traffic congestion, and the will take AVs to replace human-driven vehicles. The adop-
transportation infrastructure necessary to support a fleet of tion timeline will be a major determinant of how planners
shared vehicles. Many of the specific implications of a shared will need to respond to the emergence of AVs. Unfortunately,
vehicle system will be discussed in more detail in the follow- as with any forecasting effort, predicting AV adoption rates
ing sections of the report. This section will outline some of is a difficult task due to the host of factors that could speed
the general impacts that will trickle down into the specific or hinder adoption. AV adoption is particularly difficult to
aspects of the built environment. predict because of the rapid pace at which AV technology and
The first impact is that shared mobility could signifi- the associated regulatory framework are changing.
cantly reduce the size of the vehicle fleet. The typical privately The speed of technological change has increased over
owned vehicle in the United States is parked 95 percent of time, as evidenced by how quickly the smartphone was ad-
the day. Instead of parking, a shared AV could immediately opted and how quickly startups like Uber and Lyft made on-
pick up another passenger, thereby completing significantly demand mobility a major part of the transportation system.
more trips than a privately owned vehicle. In fact, studies While it is difficult to predict whether the same will be true of
have found that one shared AV could replace between nine AVs, the novelty, convenience, and mobility provided by AVs
and 11 privately owned vehicles (Fagnant, Kockelman, and may speed public acceptance and adoption. In fact, surveys
Bansal 2015). evaluating the public perception of AVs have shown a steady
Such a drastic reduction in the vehicle fleet could have increase in the level of trust of AVs (AAA 2018). In addition,
considerable ramifications for the built environment. Most industries such as freight and public transit may lead the way

www.planning.org AMERICAN PLANNING ASSOCIATION 21

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

because they have a greater financial incentive to transition to

FEDERAL POLICY ON AVS autonomous technology. Seeing the successful implementa-
tion of AVs in these industries may pave the way for general
The policy guidance outlined in this PAS market acceptance by calming the public’s fears over entrust-
Report focuses primarily on the response ing their safety to robots.
of local planning agencies to the adop- However, a major factor that could slow the public’s will-
tion of AVs. However, federal and state ingness to purchase AVs is the high cost of the technology.
policy will affect when and how AVs are Cost premiums for autonomous capabilities are expected
deployed and will shape local planning to be high when AVs first become available (Mosquet et al.
efforts to address AVs. 2015). While consumer surveys have shown that many are
In the report Automated Driving already willing to pay substantial premiums for fully autono-
Systems: A Vision for Safety 2.0 (NHTSA mous capabilities (Daziano et al. 2017), others may be unwill-
2017b), the U.S. Department of Trans- ing or unable to pay the extra cost, especially since the aver-
portation (DOT) provides broad policy age American already spends about 15 percent of his or her
guidance on AVs. The report is divided income on transportation costs. Consequently, how quickly
into two sections. The first section, “Vol- the cost of AVs goes down may be a primary determinant of
untary Guidance for Automated Driving when privately owned AVs are adopted. However, by taking
Systems,” provides 12 safety elements the driver out of the equation, shared AVs could significantly
to support the safe testing and imple- decrease the cost of ride sharing. ReThinkX has projected that
mentation of highly automated (Levels the per-mile transportation cost of shared AVs will quickly
3–5) technology. It is designed to help drop lower than privately owned vehicles (Airbib and Seba
AV manufacturers “identify and resolve 2017). This could lead to faster adoption of shared AVs.
safety considerations prior to deploy- Another factor that could impede the adoption of AVs
ment” (NHTSA 2017b). For each element, is how the federal and state governments regulate the use
the report provides safety goals and best of AVs. Continued support from the federal government in
practices for attaining those goals. the form of enabling regulation, mandates, or infrastructure
The second section, “Technical As- could help the process. (See the sidebar on this page for an
sistance to States,” starts by outlining overview of federal policy guidance on AVs.) However, regu-
the federal and state roles in regulating latory delays in a number of different arenas, including in-
AVs: “NHTSA is responsible for regulat- surance, legal liability, testing and validation procedures, and
ing motor vehicles and motor vehicle cybersecurity, could delay AVs’ market availability.
equipment, and States are responsible The timeline for when AV technology will be adopted
for regulating the human driver and will vary significantly based on the level of automation. Level
most other aspects of motor vehicle 1 features such as cruise control have been widely used for
operation.” From that foundation, it then some time, and many Level 2 features are already available in
provides state legislators and highway luxury cars and are likely to become widely accepted in the
safety officials with a framework of best next several years. However, while lower levels of automation
practices for developing their own laws promise notable safety benefits, they are not expected to have
and regulations. The goal of this guid- significant impacts on the form or function of the transporta-
ance is to develop a “consistent, unified tion system. Consequently, the remainder of this section will
national framework” of laws and policies focus on the adoption of fully autonomous vehicles unless
that promotes the development and otherwise specified.
implementation of AVs (NHTSA 2017b). The first determinant of AVs’ adoption timeline is when
U.S. DOT plans to release another fully autonomous vehicles will become available for sale to
update to its guidance for AVs in 2018 consumers. Designing an autonomous vehicle that can make
and is currently in the process of receiv- proper driving decisions in every situation, context, and
ing public comment to inform that doc- condition is a monumental task. OEMs are making rapid
ument (U.S. DOT 2018). progress, but the challenges of designing vehicles that can

22 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

consistently detect bicyclists and pedestrians, function in in- AVs will be adopted. Several projections for AV adoption have
clement weather, and have enough operational redundancy agreed that by 2030 AVs will constitute around 15 to 20 per-
to correct mistakes remain. cent of vehicle sales (Mosquet et al. 2015; McKinsey & Com-
Despite these challenges, many major auto manufactur- pany 2016; Walker Consultants 2017). Similarly, three studies
ers have aggressive timelines for when they anticipate offer- have projected that AVs would make up about 25 percent of
ing highly automated vehicles. Several companies, including the vehicle fleet in 2035 (Mosquet et al. 2015; Bierstedt et al.
Waymo, GM, Ford, and Volvo, have stated that they antici- 2014; Kuhr et al. 2017), while another study proposed a slight-
pate having AVs available for sale in 2020 or 2021. However, ly more conservative estimate of about 20 percent by 2040
in many cases, these aggressive timelines are regarding Level (Litman 2018). These findings strongly suggest that in just
4 automation at best. Consequently, some researchers have over 15 years, AVs could represent a quarter of the vehicles on
predicted that fully autonomous (Level 5) vehicles will not the road. This may not sound like much, but these estimates
become available for sale until around 2025 (Mosquet et al. provide important indications that it is not a question of if
2015; Underwood 2015). Yet considerable uncertainty re- but when AVs will become available. In addition, if many of
mains as more conservative scenarios that anticipate techni- the AVs on the road are shared, then AVs could account for
cal and regulatory delays predict that Level 5 automation may significantly more than 25 percent of vehicle trips. ReThinkX
not become available until after 2030 (McKinsey & Company projected that by 2030 40 percent of vehicles will be privately
2016). While these conservative predictions are the minority, owned vehicles, but they would only represent five percent of
they do underscore how difficult it is to predict an autono- passenger miles (Airbib and Seba 2017). Finally, it is a very
mous future and they highlight some of the factors that could rare and notable occurrence for a quarter of the population of
slow the development and adoption process. any community, let alone an entire country, to change their
Regardless of when AVs become available for sale, it is dominant mode of transportation in less than 15 years.
generally accepted that there will be a long transition period Unfortunately, extending these projections further into
between when the technology is introduced and when AVs the future yields greater uncertainty. Several studies esti-
reach full adoption. With approximately 260 million vehicles mate that AVs will reach about 50 percent adoption some-
in the United States and only about 17.5 million vehicles sold time between 2045 and 2055 (Litman 2018; Kuhr et al. 2017).
every year, if every vehicle sold was fully autonomous it still However, McKinsey & Company’s (2016) high-disruption
would take close to 15 years to replace the existing vehicle estimate projected that AVs will total more than 50 percent
fleet (Kuhr et al. 2017). However, due to the expected cost of the vehicle fleet long before 2040. Unfortunately, there is
premiums and consumer hesitance to trust AV technology very little literature predicting exactly when fully autono-
(Schoettle and Sivak 2014), AVs could make up a very low mous vehicles will make up 100 percent of the vehicle fleet.
percentage of vehicles sold in the early years of availability. Early forecasting studies conducted in the first half of the
Litman (2018) predicts that AVs will only make up two to five 2010s were generally more optimistic concerning the rapid
percent of vehicle sales in the 2020s and that they will not adoption of AVs. For example, a study by Morgan Stanley
make up 100 percent of vehicle sales until the 2050s. Con- (2013) projected that AVs would reach full adoption by 2035.
sequently, most researchers agree that autonomous and hu- However, as OEMs delayed their expected release date for
man-driven vehicles will share the road for decades before fully autonomous vehicles from the late 2010s to the early
100 percent of the vehicle fleet becomes autonomous. 2020s, projections have become more conservative. Very few
Due to the uncertainties surrounding when AVs will recent projections have made claims concerning when AVs
become available for consumers, when they will be afford- will reach full adoption. Consequently, a key takeaway from
able, and how quickly the public will adopt the technology, these projections is that once AV technology becomes avail-
timelines of AV adoption vary widely. This is especially true able, there will be an extended transition period of several
the further into the future the projection looks. One study decades where autonomous and human-driven vehicles will
found that fully autonomous vehicles could make up any- share the road. Yet, given the potential cost-effectiveness of
where between 10 and 90 percent of the vehicles sold in 2040 shared autonomous mobility, the percentage of autonomous
(McKinsey & Company 2016). trips could increase very quickly even if they represent a
However, shorter-term projections, such as those esti- relatively small portion of total vehicles.
mating adoption over the next 15 years, have produced fairly It is also important to recognize the role that planners
similar results and provide useful insights into how and when will play in determining the rate of AV adoption. By and large,

www.planning.org AMERICAN PLANNING ASSOCIATION 23

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

the technology is advancing much faster than the regulatory

framework AVs operate within. In this way, the regulatory
framework, or lack thereof, could govern the adoption time-
line of AV as much if not more than technological develop-
ment or market acceptance. While the regulatory framework
necessary extends beyond the control of planners to legal li-
ability, licensing, and a host of other issues, planners will play
a key role in paving the way for AVs to be allowed to be tested,
piloted, and implemented on a wide scale.

CONCLUSION

Technologies that had previously been the stuff of science fic-

tion are becoming reality and are already being successfully
tested on public roadways around the globe. If the technology
continues to advance rapidly, highly, if not fully, autonomous
vehicles are expected to become available to the public within
the next 10 years (Kuhr et al. 2017). These advanced trans-
portation technologies could have major implications for the
safety and efficiency of the transportation system. In short,
AV and CV technology is here and it is here to stay, and it is
vital for planners to begin preparing for AVs now.
The technology is evolving so rapidly that planners will
need to embrace an attitude of continuous learning. The
summary of the emerging technologies’ capabilities provided
in this chapter is merely a snapshot of where the technology
is today. The array of options available for implementation
by planners and policy makers may be radically different
in three to five years. Planners will need to monitor techno-
logical developments and successful pilot programs to stay
informed of the technologies that are available to them and
for which they need to be planning. In particular, planners
should study the successes and failures of early adopters and
their implications for long-range planning efforts in their lo-
cal contexts. These collective lessons learned will be vital in
order to avoid repeating the same mistakes.
The bottom line of this chapter is that AVs encompass a
wide range of emerging technologies that are expected to re-
shape the design of transportation systems and of our urban
spaces. While the most notable impacts will only be viable
once Level 4 and 5 AVs are the majority of the vehicle fleet,
the time to start preparing for AVs is now. Building upon this
background information regarding what the technology is
capable of and when AVs are expected to become available,
the next chapter will outline the major opportunities and
challenges AVs offer that will shape the technology’s effect on
the transportation system and on the built environment.

24 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 2

www.planning.org AMERICAN PLANNING ASSOCIATION 25

CHAPTER 3
PLANNING
OPPORTUNITIES
AND CHALLENGES
IN AN AV WORLD
As with any major advancement in transportation technology, autonomous vehicles will present planners with a set of oppor-
tunities and challenges to address as they work toward building better communities. Planners and policy makers will have the
responsibility of striving to capitalize on the opportunities while mitigating the challenges.

This chapter will lay out some of the major opportuni- able opportunities to address today’s most pressing transpor-
ties and challenges that will shape AVs’ impacts on cities and tation problems. While claims that AVs will solve all of our
the appropriate policy responses to them. Chapter 4 will then urban problems may overlook the difficult challenges the
detail the effects AVs are expected to have upon the built envi- technology will create, they are indicative of the revolution-
ronment. Finally, Chapter 5 will tie all of this back to practice ary opportunities that the technology promises to provide.
by identifying planning and policy interventions that can cap- Most notably, AVs will provide opportunities to address some
italize on AVs’ opportunities to create vibrant urban spaces. of the issues created by the automobile by
What opportunities and challenges exist for planners as
we enter an AV world? Several recent articles illustrate po- • Improving traffic safety
tential downsides of AVs and provide rationales for robust • Increasing traffic efficiency
policy responses (Riggs and Boswell 2016a; Riggs and Boswell • Reducing vehicle emissions
2016b). This report identifies several major opportunities AV • Improving mobility for special populations
technology provides to improve the form and function of our
communities and to better the lives of those living in them. This section will explore each of these potential benefits.
These opportunities include the potential to improve the
safety and efficiency of the transportation system, the ability Improved Traffic Safety
to reduce vehicle emissions, and the chance to improve the In 2015, 6.3 million automobile crashes occurred in the
mobility of transportation-disadvantaged populations. United States. Of these crashes, 94 percent were attributable
Unfortunately, AV technology also brings challenges that to human error. These crashes cost the U.S. more than $200
threaten to negate its potential benefits. In particular, AVs may billion in medical costs, property damage, traffic congestion,
reinforce auto-oriented sprawl, which could increase vehicle and lost productivity (NHTSA 2015). More importantly, traf-
miles traveled (VMT) and congestion. AVs could also com- fic accidents take the lives of more than 30,000 people every
promise bicycle and pedestrian mobility by fragmenting bi- year in the U.S. alone, with 37,461 fatalities in 2016 (NHTSA
cycle and pedestrian networks. Finally, AVs will have impor- 2017a). Worldwide there are more than 1.25 million traffic
tant ramifications on several other key planning issues such fatalities per year (WHO 2015). Motor vehicle crashes consis-
as public transit, public health, and social equity. This chapter tently rank as the number one cause of death among people
will discuss some of these primary opportunities, constraints, ages 16 to 24 (NHTSA 2016). Even beyond the loss of life, traf-
and concerns, and will offer potential policy solutions. fic accidents create substantial public health costs; for exam-
ple, such incidents resulted in more than $23.4 billion worth
of medical costs in 2010 (NHTSA 2015).
AV-RELATED PLANNING OPPORTUNITIES Technological advancements such as the seat belt and the
air bag have played a major role in improving traffic safety
All of the public attention that AVs have received in recent over time. Between 1975 and 2015, the fatality rate per 100
years is not without justification. AVs provide several remark- VMT declined from 3.35 to 1.18. Advanced driver assistance

www.planning.org AMERICAN PLANNING ASSOCIATION 27

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

systems (ADAS) and automated vehicles promise to be the lic roads and only caused one accident (Waymo 2017a). Over
next major advancement in vehicle safety technology. While the same 3.5 million miles, human-driven vehicles caused
estimates vary, ADAS such as forward collision warning and 24 traffic incidents with Google’s driverless fleet (California
automatic braking systems have been found to reduce rear- Department of Motor Vehicles 2017). Consequently, as seen
end crashes by as much as 39 percent, which equates to a 12 in Figure 3.1, Google’s total incident rate remains relatively
percent reduction in the total number of crashes and 15 per- high (about equal to the average young adult driver), but the
cent reduction in injuries (Cicchino 2016). Fully autonomous at-fault rate is significantly lower than the average driver’s.
vehicles are expected to be the next momentous innovation These early tests provide promising evidence that AVs
in transportation safety. will be able to deliver on their promised safety improve-
Since more than 90 percent of traffic crashes are caused ments; however, additional testing will be necessary to verify
by human error, removing humans from the driver’s seat has whether AVs will improve the safety of users in real-world
the potential to improve traffic safety. While AVs will not conditions. Since these companies can choose the conditions
eliminate traffic accidents, their ability to reduce or remove (weather, time of day, traffic level, etc.) in which the testing
human driving errors, such as mistakes made while drowsy, occurs, there remains insufficient data to demonstrate with
distracted, or intoxicated, may significantly reduce traffic certainty that the AVs of today are safer than human driv-
crashes and traffic-related fatalities. According to NHTSA’s ers. This is particularly true because Google and most other
traffic safety data, 28 percent of traffic fatalities in 2016 in- companies testing AVs on public roads use trained techni-
volved alcohol impairment. Another 9.2 percent of fatalities cians who take back control when the vehicle comes to a
in 2016 were caused by distracted drivers, and 2.1 percent situation it is unsure how to handle. As such, more testing
involved a drowsy driver. Since AVs do not get distracted, in- will be necessary before the AV promise of improved traffic
toxicated, angry, or sleepy, AVs are expected to significantly safety can be verified.
reduce traffic crashes and traffic-related fatalities. However, as the technology improves, the promising
While the sample of AV testing in real-world environ- safety benefits that AVs are already demonstrating are only
ments remains too small to draw definitive conclusions, early expected to increase. Waymo’s 2016 Disengagement Report
testing of highly and fully automated vehicles by Google, to the California Department of Motor Vehicles indicated that
Tesla, and others has provided promising indications of AVs’ its rate of safety-related disengagements dropped from 0.8
potential to reduce traffic accidents and traffic-related inju- per thousand miles in 2015 to 0.2 disengagements per mile
ries (Richland, Lee, and Butto 2016). Between 2009 and 2017, (Waymo 2017b). This means that a human operator was only
Google’s AVs had driven more than 3.5 million miles on pub- required to retake control of the vehicles once every 5,128

Figure 3.1. Google’s self- Crash Rate per 100 Million Miles
driving car crash rate
per 100 million miles Conventional Crash Rate (‘08/’09)
(Crow 2017)
2017 Waymo Not at Fault
Teens are extremely
crash-prone drivers 2017 Waymo at Fault

206 fewer crashes per

million miles of driving

Driver Age

28 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

miles on average, compared to every 1,244 miles in 2015, indicative of the types of problems and vehicle-to-operator
indicating a substantial improvement in Waymo’s ability miscommunications that AVs may bring during the early
to safely maneuver real-world driving situations. AVs’ traf- stages of adoption.
fic safety is expected to continue to improve because, unlike Finally, vehicle hacking may pose a new cause of traffic
conventional vehicles, AVs utilize self-learning tools so the incidents and injuries, particularly in the early years when
more miles they travel, the safer they become. This machine cybersecurity systems may not be as robust. Although not
learning can then be shared with other vehicles, making the directly related to AVs, recent cyberattacks on municipalities,
entire fleet safer over time. such as the city of Atlanta, highlight the importance of cy-
Yet in spite of all of AVs’ promise to improve driver bersecurity and how crippling a cyberattack on AVs or smart
safety, there are several safety considerations and threats that infrastructure could be. While it remains unclear how big of
planners need to be aware of, particularly as our roadways a problem hacking will be for AVs, vehicle manufacturers and
transition from primarily human-driven vehicles to AVs. local, state, and federal government agencies are putting enor-
The first is that during the transition AVs may introduce new mous effort into ensuring that adequate cybersecurity systems
safety risks. As previously alluded to, the rapid introduction are in place to prevent hackers from gaining control of AVs.
of such new and revolutionary technology is likely to bring However, even with the introduction of these new safety
with it some level of equipment malfunctions and machine risks, AVs’ abilities to improve traffic safety and save the lives
error as the technology is perfected and developers determine of thousands of people every year is the most significant ben-
how to address difficult driving situations and road condi- efit that AVs may provide and represents one of the most ex-
tions. Hopefully, most of these issues will be mitigated by the citing examples of how good planning practice can promote
extensive testing process that companies across the country public health.
are conducting, but some level of machine error is inevitable.
In addition, traffic safety risks may increase as human Greater Traffic Efficiency
drivers learn how to share the road with AVs. Drivers not In addition to the anticipated reduction in auto accidents, the
familiar with AVs’ capabilities and limitations may drive adoption of AVs is expected to improve the efficiency of the
more recklessly if they overestimate AVs’ abilities to avoid transportation system. AVs promise to positively affect traffic
collisions, which may partially offset the safety improve- efficiency and throughput in several different ways (Table 3.1).
ments provided by AVs. In some cases, these public safety First, if AVs’ expected safety benefits come to fruition,
challenges may be alleviated by dedicated infrastructure for reducing traffic accidents will also significantly reduce traffic
AVs. However, dedicated infrastructure will not be feasible congestion caused by traffic incidents. Since AVs, particularly
in every context, and drivers will need to learn to adapt to vehicles that are both connected and automated (CAVs), will
the AVs’ driving behavior. have faster reaction times than human drivers, they will be
Another safety risk may be caused by miscommunica-
tion between the AV and human operators. Many of the early
AVs are expected to possess Level 2, 3, or 4 automation in-
stead of Level 5. This would require the operator to retake TABLE 3.1. WAYS AVS MAY IMPROVE
control of the vehicle in certain situations. Unfortunately, TRAFFIC EFFICIENCY
studies have shown that when human operators are not re-
quired to pay attention at all times, they are much less likely Reducing congestion caused by traffic incidents
to remain alert enough to retake control in a timely manner
Allowing vehicles to travel closer together
when necessary (Blanco et al. 2015). The most highly publi-
cized example of this is the Tesla operator who died as a result Allowing vehicles to travel in harmony
of a traffic incident while the vehicle was in autopilot mode.
Improving throughput through intersections
A report by the National Transportation Safety Board found
that the driver “was not attentive to the driving task” because Reducing vehicle size
he had an “overreliance on the automation.” It also indicated
that Tesla’s system of monitoring the driver’s interaction was Encouraging car sharing and ride sharing
“not an effective method of ensuring driver engagement”
(NTSB 2017). Tesla has since addressed this issue, but it is (Source: Authors)

www.planning.org AMERICAN PLANNING ASSOCIATION 29

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

able to travel closer together than human-operated vehicles, were autonomous, stop-and-go waves commonly experi-
thereby increasing vehicle throughput. Studies modeling how enced on busy interstates could be significantly diminished
platooning could improve traffic throughput have found that (Stern et al. 2018).
improvements will occur gradually as AV adoption rates rise, However, AV efficiency improvements will fail to alle-
but that full market penetration of CAVs could more than viate traffic congestion and carbon emissions if AVs prompt
double vehicle throughput (Talebpour and Mahmassani significant increases in VMT. Consequently, AVs’ ability to
2016; Lioris et al. 2017). improve traffic flow will create many exciting opportuni-
Similarly, as vehicle-to-vehicle (V2V) and vehicle-to- ties, but AVs will also provide planners with the important
infrastructure (V2I) technologies improve, AVs may signifi- challenge of ensuring these improvements are not offset by
cantly improve the efficiency of intersections. Futurists have sprawling urban environments and increasing VMT.
suggested that CAVs may eliminate the need for traffic sig-
nals altogether. Instead, the vehicles would coordinate with Reduced Vehicle Emissions
each other or with the infrastructure to ensure each vehicle Closely related to AVs’ traffic efficiency benefits is their poten-
could safely pass through the intersection without coming to tial to reduce carbon emissions and improve the sustainabil-
a complete stop, potentially reducing traffic congestion by as ity of the transportation system. AVs are expected to provide
much as 60 percent (Dresner and Stone 2008; Tonguz 2011). opportunities to significantly reduce vehicle emissions, pro-
While these drastic improvements to vehicle throughput may vided the lower cost of travel and empty vehicle miles do not
represent the upper bound of potential AV impact on efficien- drastically increase VMT. Of the 6,587 million metric tons
cy of the transportation system, they certainly highlight the of the CO2 emitted in the U.S. in 2015, 27 percent was gener-
promising opportunities that AVs will provide. ated by transportation emissions (U.S. EPA 2017). More than
As previously discussed, AVs are expected to encourage half of transportation emissions are caused by passenger and
additional use of on-demand car-sharing and ride-sharing personal vehicles.
services. Increasing use of shared on-demand mobility could In addition to the negative effects these carbon emis-
improve transportation efficiency by combining trips and re- sions have on the environment, air pollution from vehicle
ducing the number of cars on the road. Studies have found emissions has been linked to heightened rates of respiratory
that a fully autonomous shared system could reduce the size disease (asthma, bronchitis), cardiovascular disease, adverse
of the vehicle fleet by up to 90 percent (Fagnant, Kockel- reproductive outcomes, cancer, and even premature death
man, and Bansal 2015). While this figure represents the up- (APHA 2009). The Federal Highway Administration has es-
per bound of fleet reduction assuming a perfectly efficient timated that air pollution from traffic generates between $50
shared-vehicle system, significant declines in vehicle owner- and $80 billion per year in health care costs (FHWA 2000).
ship are possible. However, shared mobility’s ultimate impact In short, AVs’ potential impacts to travel behavior and trans-
on throughput and congestion would depend on the extent portation efficiency could have major implications for envi-
to which the public is willing to share rides, as a car-sharing ronmental sustainability and public health outcomes in com-
model could make congestion worse by introducing empty munities across the country.
vehicle miles without reducing the number of vehicle trips. As will be discussed later in this chapter, AVs’ influ-
Finally, AV technology may improve vehicle throughput ence on travel behavior and VMT remains difficult to pre-
by reducing the size of vehicles. Vehicles are as large as they dict. However, AVs are capable of reducing vehicle emissions
are today to provide safety and versatility. Since AVs are ex- for several reasons, provided the lower cost of travel and the
pected to reduce automobile crashes and since 35 percent of empty vehicle miles do not drastically increase VMT. First,
trips can be served by one- to two-person vehicles, AVs (espe- AVs promise to significantly improve the fuel efficiency of
cially shared AVs) could be significantly smaller than today’s automobile travel. Previous efforts to improve fuel efficiency
vehicles, which would enable greater vehicle throughput. have revolved around improving the efficiency of the engine.
AVs are certainly not expected to eliminate congestion AVs introduce a new dynamic by improving fuel efficiency
caused by any of these categories, but by making notable through more efficient traffic patterns and driving behavior.
improvements in each area, the cumulative impact on traffic AVs’ ability to safely drive very close together and platoon
flow could be significant. Moreover, cities may not need to could reduce the energy consumption of road transporta-
wait for 100 percent adoption to begin to experience these tion by four to 25 percent by decreasing wind resistance. At
benefits. Even if only five percent of vehicles on the road the same time, AVs can easily incorporate fuel-optimizing

30 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

acceleration and deceleration patterns that could further re- Increased Mobility for Special Populations
duce energy consumption by as much as 23 percent (Wadud, AVs possess a special capability to restore the personal mo-
MacKenzie, and Leiby 2016). AV technology may also reduce bility of the aging and transportation disadvantaged. In to-
vehicle weight as safety features become less necessary, which day’s transportation system, aging and disabled people who
would bring even greater fuel savings. are unable to drive are often left with few transportation op-
Many researchers are also anticipating a convergence tions. As suburban populations age in place, greater numbers
of autonomous and electric vehicle (EV) technology. Several of aging adults will live in areas with limited public transit
automobile manufacturers have recently announced their in- service and poor bicycle and pedestrian infrastructure. Con-
tentions for an all-electric future in which every new vehicle sequently, when age- or health-related declines in driving
sold will eventually be an EV (Davies 2017). The added ef- ability force them to limit travel behavior or cease driving
ficiency of using electricity to power the computer systems altogether, older adults often feel trapped in their own homes
necessary to control an AV instead of converting it from (Kostyniuk, Trombley, and Shope 1998; Yassuda, Wilson,
gasoline may provide a strong incentive for more auto manu- and von Mering 1997). This process of driving cessation and
facturers to transition to electric power. It has been estimated the subsequent loss of mobility and freedom causes signifi-
that a fully autonomous fleet of EVs could reduce emissions cant deterioration to their health and quality of life. Among
by as much as 90 percent (Greenblatt and Saxena 2015). older adults, driving cessation has been found to cause higher
While it is unlikely that the impacts will be that drastic, these rates of depression, social isolation, and even mortality (Ed-
estimates demonstrate the potential for the convergence of wards et al. 2009; Marottoli et al. 1997). This will become an
advancements in AV and EV technology to work together to increasingly pressing issue as baby boomers age.
significantly reduce the transportation system’s negative en- AV technology promises to restore mobility and in-
vironmental externalities. dependence to a growing segment of the population and to
Finally, as discussed in the previous section, AVs may improve the quality of life of these populations and their
also reduce emissions by improving traffic efficiency and caregivers. Many of the existing strategies for addressing the
throughput, as AVs are expected to travel closer together, trav- issues related to driving cessation can help to mitigate the
el in harmony, and reduce accident-related congestion. If AVs mobility and quality-of-life difficulties faced by retired driv-
facilitate a drastic increase in ride sharing and car sharing, ers. However, no current strategy can provide the same level
then AVs may even reduce VMT by further amplifying the of mobility retired drivers enjoyed prior to giving up driv-
impact on vehicle emissions. However, as will be discussed ing, regardless of where they live or how strong their sup-
further in the next section, each of these effects could eas- port network is. The emergence of AV technology may be the
ily be offset if AVs substantially increase travel demand. AVs first initiative with the potential to provide older adults with
pose a major risk of drastically increasing VMT by reducing personalized rapid transit. In particular, AVs could help to
the perceived cost of travel, by improving mobility for those improve the mobility of the growing number of older adults
who previously were unable to drive, and by introducing who reside in suburban and rural areas and to minimize risks
empty vehicle miles as the vehicle travels without a passenger associated with aging drivers. A 2003 study found that 79
to find parking, return home, or find another passenger. percent of adults age 65 and older live in car-dependent sub-
Consequently, planners’ ideal roles regarding AVs’ im- urban and rural communities, which typically require fre-
pacts on emissions will remain similar to what they have quent, long-distance trips by automobile (Rosenbloom 2003).
been since the introduction of the automobile. Planners will Since providing public transit to these areas is extremely dif-
need to continue to seek creative ways of addressing traffic ficult, AVs may be a better way of ensuring older adults can
congestion, promoting the use of transit and active modes of maintain their quality of life.
transportation, mitigating potential increases in VMT, and Even Levels 1 and 2 of automated technology could serve
promoting the development and use of EVs. AVs will simply to keep aging drivers behind the wheel longer by reducing the
add the new challenges of minimizing empty vehicle miles risk of crashing and driving-related anxiety in stressful situ-
and preventing AVs from escalating urban sprawl farther ations (Duncan et al. 2015). For drivers who have completely
into rural areas. Yet, AVs’ potential efficiency improvements stopped driving, a fully autonomous vehicle offers complete
will provide important opportunities to reduce transpor- use of the automobile and restores their independence. AVs
tation emissions and improve the health of communities could serve to improve and extend older adults’ quality of life,
across the country. independence, and mobility, as well as to reduce their likeli-

www.planning.org AMERICAN PLANNING ASSOCIATION 31

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

hood of being admitted to long-term care facilities. AVs pro- 20th century caused massive changes to land-use patterns by
vide a strategy that can potentially accommodate the travel enabling people to live significantly farther away from central
behavior of aging populations within the context of a pre- cities and employment centers. The suburban development
dominately auto-focused transportation system. patterns and edge cities made possible by the automobile still
AVs may also provide opportunities to improve the mobil- dominate the urban form of most American cities today. As
ity of lower-income and transit-dependent populations. While AVs represent the most significant advancement in personal
the higher costs of purchasing an AV may make vehicle own- mobility since the mass production of the automobile, AVs
ership more difficult, the potential cost-effectiveness of shared will certainly have a dramatic impact upon the urban fab-
AVs could provide additional mobility options to transit-de- ric. AVs’ impacts on the cost and ease of transportation will
pendent populations. As further discussed later in this chap- inevitably affect the location decisions of both residents and
ter, there will also be opportunities to integrate shared-mobil- businesses. However, there is an ongoing debate over whether
ity applications into the transit system as first- and last-mile AVs will spark another wave of urban sprawl or whether they
connections. Especially if trends toward the suburbanization would prompt the reurbanization of urban centers.
of poverty continue, shared AVs may be an effective way of The majority of researchers contend that AVs have the
providing access to transit systems and meeting the mobil- potential to induce sprawl by encouraging people to move
ity needs of lower-income households. As will be discussed farther away from urban centers. By removing the respon-
in more detail later, AVs may also provide new challenges to sibility of driving, AVs will make traveling much more en-
transit-dependent populations, but it is important for planners joyable and less stressful. People may be willing to commute
to recognize and work toward the potential benefits. significantly farther if they are able to sleep or be productive
during the trip. In addition, if AVs improve the efficiency of
the transportation system, then commuters may be able to
AV-RELATED PLANNING CHALLENGES travel farther in the same amount of time. The average com-
muter typically is unwilling to commute much more than an
Like any disruptive technological improvement, the rise of hour (Kung et al. 2014). By increasing the distance commut-
AVs will bring an entirely new set of challenges and diffi- ers can travel in an hour, AVs could create pressure to push
culties that planners will have to navigate to work towards suburban development farther into previously rural areas.
building better communities. In fact, many of AVs’ most dif- Some have also suggested that AVs may encourage
ficult challenges could offset the technology’s most notable sprawl by reducing the monetary cost of travel (Burns, Jor-
benefits, creating the opposite effect. For example, in spite of dan, and Scarborough 2013; Litman 2018). This likely will
the potential efficiency benefits of AVs, they could ultimately not be true in the near term, because AVs’ sensors and com-
lead to more congestion if AVs significantly increase VMT. puter systems will raise vehicle costs. Yet, if AVs reduce
Consequently, planners have the responsibility of determin- car ownership by sparking a rise in on-demand automated
ing how to use the opportunities the technology provides to mobility, they may reduce travel cost per mile because the
address current planning problems and to proactively avoid traveler does not have to purchase or maintain the vehicle.
the problems the technology may create. While it remains to be seen whether automated car shar-
This section will describe a few of the planning-related ing will become a popular model and whether on-demand
challenges that AV technology may present. The widespread AVs would cost less than owning an automobile today, lower
adoption of AVs will also have massive implications for pri- monetary cost would encourage an intensification of sprawl.
vacy concerns, insurance, legal liability, and cybersecurity. In this way, AVs’ potential reduction of the generalized cost
While these issues may affect when and how AVs become of travel (time, stress, and money) may make people more
available, planners will not have an influence on shaping willing to live further away from central cities, greatly ex-
how these issues will be resolved. Consequently, this section panding already sprawling communities.
will not address these issues but will instead focus on AVs’ However, many proponents of AV technology suggest
planning-related challenges. that the technology will provide opportunities for promot-
ing more compact development patterns. Recent urbaniza-
Potential to Reinforce Auto-Oriented Sprawl tion trends and lower rates of car ownership among younger
There is little debate that the increase in personal mobility generations have been well documented. A shared-AV system
provided by the introduction of the automobile in the early could reinforce these growing trends by enabling people to

32 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

discard their private vehicles and move into more walkable What used to be a stressful commute to work may turn into
city centers that are better served by shared-AV systems. the opportunity to nap, read, or get more work done. Con-
Since it would be easier to provide on-demand AV service to sequently, two-hour commutes and long road trips may be-
densely populated areas, the convergence of AV technology come more common as people become willing to travel far-
with the rise of the sharing economy could serve to improve ther and more often.
the accessibility of urbanized areas while further reducing Beyond lowering perceived costs, AVs may also reduce
the cost of travel. By providing better and more affordable the monetary costs of traveling. According to researchers at
service in urbanized areas, AV could provide additional fuel Columbia University’s Earth Institute, a fleet of shared AVs
for the recent urbanization trends. could reduce taxi travel cost from $4 per mile to 50 cents
However, it is unlikely that either of these of scenarios per mile (Burns, Jordan, and Scarborough 2013). If AVs re-
would occur exclusively. Even today, urbanization and fur- duce the cost of travel this drastically, it could significantly
ther suburbanization are occurring simultaneously in dif- increase travel demand. Yet, whether AVs will provide these
ferent parts of our metropolitan regions. AVs may make cost savings remains up for debate. Researchers at the RAND
the urban core (and urban lifestyle) more attractive to some Corporation claimed that a shared-AV system would elimi-
while making the exurbs and rural areas more attractive to nate the fixed costs of car ownership, but would also increase
others. Ultimately, planners will need to be prepared for both the cost per trip, leading to an overall reduction in vehicle
scenarios to happen simultaneously. It is conceivable if not travel (Anderson et al. 2016).
likely that AVs will make urban living even more attractive to AVs could also affect travel demand in several other
younger generations while also enabling the remainder of the ways unrelated to the cost of travel. First, AVs may increase
population to move farther and farther into rural suburbs. travel demand by offering independent mobility to non-
Yet, AVs’ potential to reinforce and amplify suburban sprawl drivers, such as children, the elderly, and the disabled (Harb
will pose a real threat to the vibrant urban communities that et al. 2018; Litman 2018). Nondrivers have been limited to
planners strive to create. alternative modes of transportation, which often inhibits
their travel behavior. The improved mobility provided by
Potential for Increased VMT AVs may prompt a substantial increase in travel among
and Vehicle Emissions these special populations.
As discussed earlier in this chapter, AV technology will pro- Potentially one of the most significant increases in vehicle
vide notable opportunities to improve the efficiency of vehicu- travel may come from empty vehicle trips made by AVs be-
lar travel as AVs are expected to increase traffic efficiency and tween passenger trips. Fully autonomous vehicles will be ca-
throughput, increase fuel efficiency, and improve the viability pable of traveling without a passenger to return home while
of EVs. Yet, whether these efficiencies will lead to a decline in not in use, to find parking, or, in the case of shared AVs, to pick
congestion and total emissions will ultimately depend on how
AV technology affects travel demand. If AVs significantly in-
crease VMT, congestion and vehicle emissions may continue
to rise despite the improvements in efficiency (Table 3.2). TABLE 3.2. POTENTIAL AV IMPACTS ON
While AVs’ ultimate impacts on travel demand are not VEHICLE EMISSIONS
as clear as their effects on safety and efficiency, AVs are likely
to affect travel demand in several ways due to their effects on Ways AVs Could Reduce Emissions Ways AVs Could Increase Emissions
development patterns and the cost of travel. Urban sprawl • More efficient driving/energy • Increasing traffic congestion
has always been closely associated with increasing VMT. As use (platooning, improved fuel • Increasing VMT (more driving,
people and jobs move farther away from central cities, they economy, lighter vehicles) zero-occupancy vehicles, ve-
• Reducing traffic congestion hicles cruising or double-park-
must drive farther to reach their destinations, creating traffic
(fewer accidents) ing instead of paying to park)
congestion and increasing carbon emissions. Consequently,
• Reducing VMT (if paired with
if AVs promote sprawling development patterns, as dis- ride sharing and car sharing)
cussed in the previous section, they may also increase VMT • Convergence with electric
and amplify congestion. vehicles
In addition to the impact on development patterns, AVs
are generally expected to lower the perceived cost of travel. (Source: Authors)

www.planning.org AMERICAN PLANNING ASSOCIATION 33

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

up another passenger. It remains unclear how much empty ve- tation. The adoption of AVs promises to improve the safety
hicle trips will increase VMT, but it likely will be substantial. and efficiency of the vehicular system, but AVs’ ultimate ef-
AVs may also stimulate demand from new types of ve- fects on bicyclists and pedestrians are difficult to predict. As
hicles, such as those designed specifically to deliver restau- will be discussed further in the next chapter, AVs may require
rant food, groceries, or clothes. Amazon has made headlines less space than human-driven vehicles as car ownership de-
with its Prime Air drone delivery service, but automated creases, lane widths become narrower, road diets become
ground delivery vehicles may also grow into a notable por- more common, and the need for on-street parking is reduced
tion of VMT as unmanned delivery trips increase. Since (Litman 2018; Airbib and Seba 2017). This could provide op-
AVs could reduce shipping costs by removing the need for portunities to retrofit vehicular infrastructure such as lanes
a driver, demand for the delivery of basic goods needed on a and parking into bicycle and pedestrian facilities.
daily basis may increase. On-demand delivery services (Uber However, the emergence of AVs could ultimately re-
for goods and packages) are already growing in popularity, inforce an auto-oriented transportation system. Especially
especially in big cities. If AVs promote greater use of these given the novelty of AVs, it would be very easy to give prior-
services to deliver groceries and retail goods, this could add ity to AVs instead of designing urban spaces on principles of
another growing source of VMT. Urban VMT growth could human-centered design. Careful planning will be required
be mitigated if shared AVs transported passengers and deliv- to ensure that investments in AV infrastructure do not frag-
ered goods simultaneously, but an increase in the number of ment bicyclist and pedestrian networks. AVs’ potential to re-
semi-trucks making long-distance hauls could still increase move the need for traffic signals is a telling example of this.
VMT on the highways. Free-flowing intersections would provide massive improve-
In short, AVs are expected to increase travel demand ments to traffic efficiency from the viewpoint of vehicular
and VMT. However, there is no consensus on whether AVs transportation. However, without pedestrian signals or ma-
will ultimately alleviate or exacerbate traffic congestion and jor investments in above- or below-grade bicyclist and pedes-
vehicle emissions. Due to the lack of data on real-world test- trian infrastructure, free-flowing intersections could become
ing of AVs, estimating their ultimate impact on emissions, major barriers to bicyclist and pedestrian connectivity.
throughput, or congestion remains a difficult task. The few Ultimately it will be up to planners to balance these is-
attempts that have been made have produced wide-ranging sues and to develop context-specific, fiscally responsible so-
results (Kockelman et al. 2017; Friedrich 2016). A study by the lutions that leverage the ways AV technology could enhance
National Renewable Energy Laboratory found that energy traffic flow without compromising bicycle and pedestrian
use could decrease by as much as 90 percent or increase by as travel. In this way, promoting the use of active modes will
much as 250 percent depending on a wide variety of factors remain an important priority for planners to support the de-
including ownership model (private versus shared), EV use, velopment of healthy and sustainable communities.
and development patterns (Brown, Repac, and Gonder 2013).
More testing will need to be done to determine whether
AVs’ improvements in traffic efficiency will outweigh the as- OTHER SECONDARY IMPACTS OF AVS
sociated increase in VMT. However, decades of planning his-
tory have taught us that simply increasing roadway capacity, AVs’ impacts on traffic safety and efficiency have received the
as AVs promise to do, has consistently failed to relieve traffic most public attention, but their impacts on travel costs, pref-
congestion due to latent demand. Consequently, it is possi- erences, and patterns will have significant ripple effects on a
ble if not likely that even if AVs do significantly increase the host of other issues. The impacts to development patterns and
throughput of existing infrastructure, the combination of the the built environment will be discussed in the next chapter,
potential increase in VMT and latent demand would prevent but this section will detail a few of the additional impacts to
AVs from reducing traffic congestion and could threaten to the transportation system and our communities’ quality of
make traffic congestion even worse than it is today. life, including transit systems, public health, and equity impli-
cations. In all three of these cases, AVs could ultimately have
Potential Impacts to Active either a positive or negative effect. Ultimately, whether these
Modes of Transportation changes will improve transit service, improve public health,
Another pressing question is whether AVs will ultimately or mitigate transportation equity issues will depend on how
promote or discourage the use of active modes of transpor- planners prepare for and respond to the adoption of AVs.

34 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

AV Impacts on Transit Systems and hidden costs of owning and operating an automobile has
Automated vehicle technology offers a number of excit- made it extremely difficult for transit agencies to attract rid-
ing opportunities to improve the coverage and efficiency ers who live farther than a quarter- or half-mile away from a
of transit service. Yet as in so many other aspects of the transit stop. The automobile’s ability to provide door-to-door
technology’s impact, it brings some disruptive possibilities travel to a destination make the “first mile” from home to the
along with the opportunities. Most researchers and prac- transit stop and the “last mile” from the transit stop to a final
titioners agree that AVs will only have a limited influence destination a major contributor to low ridership. Whether the
on high-capacity rail systems, because even if AVs could first- and last-mile gaps can be filled by a fleet of shared AVs
provide affordable door-to-door service to everyone, the or by low-capacity autonomous shuttles remains to be seen.
roadways may not have the capacity to accommodate the However, the expected cost savings and the potential for low-
additional trips. However, the introduction of AV technol- capacity (10–12 passengers) feeder shuttles promise to pro-
ogy could have a significant effect on the roles and service vide opportunities to expand coverage beyond major roads
models of bus networks and paratransit. into neighborhoods. Some have even suggested that AVs will
The first and most exciting opportunity that AV tech- transform the dominant transit model from a fixed-route sys-
nology will provide transit agencies is the potential to sig- tem to personal rapid transit, in which fleets of low-capacity
nificantly reduce operational costs by removing a large part AVs provide personalized or even door-to-door transit ser-
of labor costs. Labor costs typically are one of the largest vice (McKinsey and Company 2017). While such a revolu-
components of a transit agency’s operating budget. By re- tionary transformation may be less likely, AVs offer oppor-
ducing the need for bus operators, transit agencies will be tunities to reinforce the importance and viability of public
able to operate much more cost effectively. Some of these transit by providing millions more potential riders with easy
cost savings will be offset by the higher capital costs neces- access to transit stops.
sary to purchase autonomous buses, but a recent estimate The rise of shared AVs and the growing popularity of
found that an autonomous bus could provide more than transportation network companies (TNCs) such as Uber
$3 million in savings over a 12-year bus’s life cycle, even and Lyft could have another transformative impact on the
after accounting for the increase in capital costs (Quarles public transit system by having large private companies be-
and Kockelman 2018). Such substantial cost savings could come an important part of the transit network. It remains to
be used to make significant improvements in service by in- be seen how that would play out, but there may be significant
creasing route coverage or frequency. opportunities for public-private partnerships to provide the
In addition, the safety improvements provided by au- best possible transit service to all residents. However, this
tonomous buses could also improve the cost-effectiveness also presents the danger of relying too heavily on TNCs to
of transit service by reducing the costs associated with traf- meet the mobility needs of transit-dependent populations.
fic accidents and legal liability. A recent study found that Significant equity issues could be created if local govern-
the average transit agency spends an average of 3.9 cents per ments begin to redirect resources away from transit systems
rider on liability claims and the cumulative cost of claims because they assume TNCs will meet the needs of transit-
has been going up (Aon Risk Solutions 2016). This may not dependent populations.
sound like much, but it adds up. The Metropolitan Trans- Another way AV technology could significantly alter
portation Authority in New York City paid more than $1.1 the transit model prevalent today is by reducing the need for
billion in liability claims between 1996 and 2007, and it es- paratransit service. Since paratransit is designed for those
timated that an additional $1.2 billion had been filed but unable to use traditional transit systems, it primarily serves
not yet paid (DiNapoli and Bleiwas 2008). AVs will certainly those who are also unable to drive. As previously discussed,
not eliminate these costs, but combining incident savings AVs promise to restore personal, independent mobility to
with operational cost savings provides possibilities for tran- these special populations by removing the need to be physi-
sit agencies to significantly expand their bus transit services cally able to drive. As such, many paratransit riders may sim-
without increasing their budgets. ply be able to use a personal AV. In addition, if transit agen-
The first- and last-mile problem has been one of the big- cies move toward point-to-point service models, as will be
gest barriers to increasing transit ridership in the U.S. The discussed later, then a separate paratransit service may not
combination of sprawling suburban development patterns, be necessary because the predominant transit model would
fragmented bicycle and pedestrian infrastructure networks, function much like paratransit does today.

www.planning.org AMERICAN PLANNING ASSOCIATION 35

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

Having paratransit operate as an integrated part of the The technology could also provide opportunities to address
larger system could improve the quality and efficiency of ser- a host of other health issues associated with the automobile,
vice to disabled populations. This is especially relevant for such as air pollution, healthy aging, changes to active trans-
transit agencies because paratransit and shuttle services are portation, and commuter stress (Crayton and Meier 2017).
the most expensive modes of transportation for any public However, just like the automobile, AVs will likely be “accom-
agency to operate. In this way, automated vehicles could serve panied by a plethora of unintended consequences” that may
to further reduce costs while expanding services to the rap- threaten the health, safety, and welfare of drivers and non-
idly growing demographic of aging citizens. drivers alike (Richland, Lee, and Butto 2016). Many of these
Finally, the positive public attention accompanying the issues, listed in Table 3.3, have been discussed in previous sec-
novelty of AV technology may provide opportunities for new tions but they are worth highlighting again here because of
funding sources and marketing opportunities during the the significant impacts AVs may have upon health outcomes.
transition to autonomous buses. Federal and state govern- In summary, AVs offer several exciting opportunities to
ments have already begun offering numerous funding op- improve the public health of communities across the country.
portunities for innovative transportation technologies. The More specifically, AVs may:
U.S. Department of Transportation and the Department of
Energy are allocating millions of dollars toward the develop- • Significantly reduce injuries and fatalities caused by traffic
ment and implementation of smart-city and connected ve- incidents
hicle (CV) technologies. In particular, the U.S. Department • Improve air quality and decrease pollution-related rates
of Transportation allocated $160 million to the Smart Cities of respiratory disease, cardiovascular disease, and cancer
Initiative in 2015 to support communities’ efforts to develop • Improve health and quality-of-life outcomes of aging and
infrastructure that harnesses the growing availability of data disabled populations by increasing personal mobility
to improve the lives of residents (White House Office of the • Relieve the mental toll of commuter stress
Press Secretary 2015). Similar funding sources may become
available for local transit agencies to pilot and implement au- Many of these opportunities may not come to fruition
tomated transit systems. For example, the Federal Transit Ad- until fully autonomous vehicles make up the majority of the
ministration (FTA) announced the Strategic Transit Automa- vehicle fleet. Yet, in the near term, policy makers and auto
tion Research Plan that will promote transit applications of manufacturers can work to ensure that as many cars as pos-
automated technologies through enabling research, integrat- sible are equipped with emergency braking and other ADAS
ed demonstrations, and strategic partnerships (FTA 2017). As that have been shown to prevent traffic incidents.
part of this program the FTA has committed funds to support
Valley Metro’s shared-AV pilot in Phoenix (FTA 2017). These
funding sources may only be available in the near term but
they provide excellent opportunities for communities to le- TABLE 3.3. POTENTIAL HEALTH IMPACTS OF AVS
verage the funding to improve existing service.
Similarly, the public attention surrounding AVs will Expected Net Health
likely bring novelty riders to use public transit for the experi- Impact
Impact
ence of riding in an AV. Granted, once the novelty wears off,
mode choice decisions will still be determined by cost, con- Traffic safety +
venience, and commute time. Yet, during the transition to an Air quality (emissions) +/-
automated future, local transit agencies can use the positive
Reduction in travel stress +
public attention to find creative ways to market and promote
long-term ridership. This could help to overcome the negative Use of active modes of transportation +/-
perception of public transit.
Improved mobility for transportation-
+
disadvantaged populations
Impacts on Public Health
AVs’ potential to save lives lost in traffic accidents is the most Health equity -
commonly cited benefit this technology could provide. Yet,
AVs’ impact to public health reaches well beyond traffic safety. (Source: Authors)

36 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

However, AVs’ impacts on travel behavior and urban fail to fully resolve the transportation problems faced by dis-
form will have important implications for how the technol- abled populations. As such, aging and disabled populations
ogy affects our communities’ health and well-being. In fact, often experience social isolation and feel trapped in their own
if AVs promote further sprawl and drastically increase VMT, homes (Yassuda, Wilson, and von Mering 1997). In addition,
the new technology could have several negative health exter- as already discussed, the relative expense and inefficiency of
nalities, including: paratransit puts a considerable burden on transit agencies
trying to meet the mobility needs of the aging and the dis-
• Lowering air quality abled. By promising to restore personal mobility to the blind
• Reinforcing sedentary lifestyles by promoting inactive and otherwise disabled, AVs could solve both of these prob-
modes of transportation lems at the same time. In this way, AVs present the potential
• Widening the health divide between wealthy and poor for an exceptional advancement in equitable transportation.
communities Unfortunately, AVs’ improvements to personal mobility
may only be available to those who can afford them. It is too
The importance of careful planning is highlighted by the soon to predict whether AVs will raise or lower the cost of
fact that many of these potential externalities are the exact driving in the long run. However, in the early stages of the
opposite of AVs’ possible health benefits. Planners will need transition from human-driven to automated vehicles, AVs are
to ensure that “decisions about AVs are made in the context expected to be significantly more expensive than traditional
of AVs’ overall impact on society” (Richland, Lee, and Butto automobiles due to the high costs of the sensors, Lidar, and
2016), as AVs’ influence on travel behavior and urban form computer systems necessary for an AV to operate. Conse-
will have important implications for our communities’ health quently, the costs of owning a car will likely go up, making it
and well-being. Planners will face the challenge of promoting more difficult for lower-income populations to maintain their
the use of AVs in a way that provides the expected safety ben- personal mobility. As the technology advances and becomes
efits without causing a host of other health issues. more pervasive, AVs’ purchase prices will likely decline. Yet,
it is unlikely that AVs will ever cost as little as vehicles today.
Social Equity Impacts Since lower-income families already spend a higher propor-
Like many other disruptive technologies before them, AVs tion of their income on transportation than wealthier house-
have the potential to mitigate or exacerbate numerous social holds, increasing the cost of car ownership could become
equity issues. Unfortunately, even though most researchers problematic for many low-income families to struggling to
acknowledge that AVs will have notable ramifications for make ends meet (FHWA 2014).
social equity, particularly as it relates to equitable access to However, due to the high costs of AV ownership and the
resources, jobs, and amenities, there has been a very limited emergence of the sharing economy, many believe that AVs
amount of research conducted on the equity impacts of AVs will encourage greater use of shared on-demand mobility
(Milakis, van Arem, and van Wee 2017). While it is too soon systems. By placing the cost of purchasing the vehicle on the
to determine who will benefit most from AVs and whether service provider instead of the individual, shared AVs could
AVs will improve or hinder access to affordable mobility, it is significantly reduce the cost of transportation to the user.
vital for planners to begin considering the equity implications This would still be profitable for the service provider because
sooner rather than later to ensure that the safety and mobility the vehicle would be running almost continuously instead
benefits of AVs are not only for those who can afford them. of sitting in a parking lot or a garage most of the day. Users
One of the most commonly cited potential positive equi- would only pay for the cost of the miles they travel, which
ty impacts of AVs is the improved mobility that autonomous could reduce the per-mile cost of travel. Estimates vary on
technology could provide to transportation-disadvantaged how much travel costs could be reduced, but most agree that
populations such as the young, the aging, and the disabled. shared AVs may likely become the cheapest form of motor-
This has been discussed previously, but it is worth mention- ized travel (Burns, Jordan, and Scarborough 2013; Johnson
ing again as the lack of viable transportation options for those 2015; Bosch et al. 2017). In this way, the widespread use of
unable to drive a car has become one of the most pressing shared AVs would help to reduce one of the largest barriers
transportation equity issues of our time. While ADA require- preventing the economically disadvantaged from enjoying
ments and paratransit help to mitigate accessibility issues and the mobility benefits of the automobile and could greatly im-
provide mobility options to those unable to drive, they often prove access to economic opportunities. However, it remains

www.planning.org AMERICAN PLANNING ASSOCIATION 37

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

to be seen whether a shared-AV system will rise to promi- dustries, it is uncertain how AVs will ultimately affect total
nence as the dominant ownership model or whether a shared employment. However, it seems likely that AVs will reinforce
system would drive down transportation costs by as much as the growing mismatch between jobs and skills as more and
it is predicted. more blue-collar jobs are replaced by white-collar tech jobs.
Historically, transportation inequalities have centered Finally, some have expressed concerns that the attention
around mobility and accessibility and whether everyone surrounding AVs may “serve as another distraction from the
has access to affordable mobility options. However, if the urgent need to design (or redesign) cities for people” (Ko-
safety benefits of AVs are only affordable to wealthy popula- dransky 2016). The energy and resources given to develop-
tions, then AVs could introduce inequalities in transporta- ing the infrastructure and legal framework for AVs could be
tion health and safety outcomes. Although AVs will likely put toward solving more pressing issues such as the need for
improve the safety of all road users, human-driven vehicles affordable housing, access to healthy food, environmental
will likely be responsible for a greater percentage of traffic ac- justice, and the creation of vibrant urban spaces where all
cidents. In this way, transportation could become more dan- people have access to urban jobs and amenities. While it is a
gerous for those unable to afford an AV, making safety while false dichotomy to suggest that adopting AVs and addressing
driving functionally dependent on income, which opens up a these important equity issues are mutually exclusive goals,
host of difficult equity and public safety issues. these concerns do raise the important point that AVs will
The transformative impacts of AV technology will ex- reinforce the auto-dominated U.S. transportation system
tend far beyond the operations of the transportation system and all of the problems that come with it. Consequently, it
to the business models of the transportation industry and will require a concerted effort to prevent AVs from exacer-
the economy at large. While this transformation offers some bating many of the urban development patterns that led to
promising opportunities, it will also have some disruptive ef- the social exclusion, spatial (job/housing) mismatch, lack of
fects that could negatively impact specific interest groups. In multimodal mobility, and other equity issues faced by many
particular, automated technology’s ability to replace the driv- low-income families today.
er may be a relief to the average commuter, but it poses a threat One example of this is AVs’ potential to pull resources
to the livelihood of motor vehicle operators such as truck, bus, away from public transit. Some have suggested that the re-
and taxi drivers. A recent study by the Department of Com- sources and funding necessary to enable a smooth transition
merce found that the introduction of AVs could affect 15.5 to AVs could leave less funding for public transit (Richland,
million workers (Beede, Powers, and Ingram 2017), including Lee, and Butto 2016). This neglect of public transit could be
3.8 million motor vehicle operators who could lose their jobs detrimental for those lacking access to other forms of trans-
as a result of AV technology. The other 11.7 million include portation. However, as discussed previously, AVs will also
construction workers, waste management professionals, first provide notable opportunities to improve transit systems by
responders, health care professionals, and others where driv- reducing operating costs and providing a first- and last-mile
ing is a major part of the job. This larger group may not lose solution. Realizing these goals will take a concerted effort by
their jobs, but they may still experience significant changes planners and policy makers to ensure that the resources put
due to the growing use of automated technologies. toward promoting AVs do not simply go toward private AVs
Suggesting that AVs will replace millions of jobs may be a but toward incorporating AVs into the design of great urban
little misleading because it does not account for the jobs that a places that provide all residents with opportunities and ac-
potentially trillion-dollar industry may create (Lanctot 2017). cess to jobs, service, and amenities.
In addition to the information technology opportunities that In summary, whether AV technology will improve or ex-
will continue to develop, AVs will likely transform the role of acerbate transportation equity issues will depend on whether
motor vehicle operator jobs from operator to support. Par- AVs reduce or increase the cost of travel. It is too early to de-
ticularly in the early stages of implementation, many AVs will termine exactly how AVs will affect the cost of transportation
likely need an operator to retake control of the vehicle in case and whether AVs will improve access to affordable mobility.
of a system failure. Freight operators may still be necessary to A shared-AV system offers exciting opportunities to reduce
handle vehicle maintenance and delivery. Automated buses transportation costs and improve access to reliable trans-
will likely still need an operator to act as a security guard and portation, thereby improving economic opportunities and
to assist disabled riders in using the system. While it is clear outcomes. Yet AVs also threaten to reinforce several grow-
that AVs will have a transformative effect on numerous in- ing equity issues by impacting millions of jobs, reinforcing

38 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 3

development patterns that disadvantage those with limited

transportation options, and impacting service to transit-de-
pendent populations. It is vital for planners to be mindful of
AVs’ equity effects to ensure our cities are designed for people
and not for AV technology.

CONCLUSION

Too often the conversation around AVs envisions either a

utopian future where the technology will neatly solve many
of the problems planners have been trying to address for de-
cades, or a dystopian future where AV technology exacerbates
urban sprawl to the point of gridlock, speeds climate change,
and creates major cybersecurity and data privacy problems.
While recognizing the potential for both of these extreme fu-
tures to be useful, it is vital to keep in mind that the technol-
ogy itself will not bring about either of those scenarios. The
auto-oriented suburban development patterns that dominate
our cities today were not created solely by the adoption of the
automobile. The technological transformation was support-
ed by subsidized infrastructure, a booming economy, and a
lack of growth management. The ultimate impact of AVs will
largely be determined by infrastructure and policy decisions
made by local governments across the country.
Consequently, we planners are better served by framing
our approach to AVs around the recognition that, like any
major technological innovation in transportation, the rise of
AVs will bring a new set of opportunities and challenges that
we will have to navigate in order to work toward building bet-
ter communities. As such, we have the responsibility to deter-
mine how to use the opportunities the technology provides
to encourage the utopian future instead of the dystopian. In
other words, there is a growing need for planners to develop
policy and infrastructure solutions that enhance the societal
benefits that AV technology promises to provide while miti-
gating the potential problems the technology could create. To
that end, this report hopes to provide planners with infor-
mation and guidance on how to begin planning and prepar-
ing for the adoption of AV technology. To do this, planners
need to understand the range of potential built environment
changes AVs may bring. The following chapter will detail
possible impacts to the design and functioning of the built
environment and will provide a starting point for state and
local agencies to consider how best to leverage these effects to
improve communities across the country.

www.planning.org AMERICAN PLANNING ASSOCIATION 39

CHAPTER 4
POTENTIAL
IMPACTS OF AVS
ON THE BUILT
ENVIRONMENT
It is clear that autonomous vehicles are poised to revolutionize the way people travel and to have a profound impact on the
built environment (Chapin, Stevens, and Crute 2017). While there is uncertainty surrounding the exact form these effects may
take, there are numerous ways that AVs will affect the way planners prepare for an AV future. These impacts include smaller
roadways, more efficient rights-of-way usage, access management changes, signage and signalization changes, and pedestrian
and bicycle interface effects, as well as parking reduction and location changes. These changes will also present significant
redevelopment opportunities. Each of these effects is explored in the following sections.

Exactly how and when AVs will impact the built envi- made possible by a number of factors. First, autonomous ve-
ronment is difficult to predict due to the uncertainty sur- hicles are expected to be smaller with the ability to drive more
rounding major factors, including the size and design of AVs, precisely (Fagnant and Kockelman 2014; Litman 2018). Even
anticipated changes to the vehicle ownership model (from when drivers try their best to keep their vehicle in the center
private ownership to shared), the cost of AVs, and their ob- of the lane they inevitably will move back and forth within
served impacts on development patterns and vehicle miles the lane. Consequently, roadways today are designed to pro-
traveled. This chapter presents a vision of the future that il- vide space to move side-to-side without crossing into an adja-
lustrates some of the built environment challenges and op- cent lane and putting other vehicles at risk of a collision. AVs
portunities that may arise with the transition to an AV fleet in are expected to remove the need to design roads and lanes
the coming decades. It takes a longer-term view of the impact to account for human error. AVs’ ability to move more pre-
of AVs on the built environment, as many of the impacts dis- cisely than human-operated vehicles will enable roads to be
cussed are only possible once AVs constitute most or all of the designed with narrower lanes.
vehicle fleet. Exactly how much lane widths could be reduced will de-
As there likely will be a long transition period where pend on how AVs will be designed (i.e., how wide AVs are). It
AVs share the road with human-driven vehicles, built envi- has been suggested AVs will be smaller than cars today, which
ronment changes during the transition are also given some could enable substantial width reductions once 100 percent
attention. However, the main focus of this report is on the of the vehicle fleet becomes automated. During the transition
long-term effects of AVs, which provides some direction for to a fully automated fleet, narrower traffic lanes may only be
planners as they review and update their communities’ long- possible in dedicated AV lanes, which could be designated
range plans. The impact of AVs is likely to vary based on the and striped much like high-occupancy vehicle lanes are now.
community context (i.e., different impacts in urban cores and But as AVs are adopted, all roadways may be designed with
rural areas), and we have chosen to focus primarily on AVs’ narrower lanes, which will leave more space for bicycle and
impact on urban and suburban areas. pedestrian facilities, active streetscapes, or green spaces.
Second, AVs offer the potential for increased throughput
(Anderson et al. 2016). As was discussed in Chapter 3, AVs
RIGHTS-OF-WAY are expected to improve the efficiency of automobile travel
by reducing congestion caused by crashes, enabling vehicles
Widespread adoption of AV technology will likely have a sub- to travel closer together, and improving traffic flow through
stantial impact on street design, with the potential for nar- intersections due to the emergence of free-flow intersections.
rower pavement widths and more efficient vehicular rights- When combined, these factors could significantly improve
of-way. Slimmed-down pavement and rights-of-way will be the vehicular throughput of existing roadways, although

www.planning.org AMERICAN PLANNING ASSOCIATION 41

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

that could have negative consequences for bicyclists and public right-of-way by reducing the number of auto travel
pedestrians. The space needed to accommodate auto traffic lanes and narrowing the width of travel lanes to less than
could be reduced without negatively affecting throughput the normal 11 or 12 feet. In fact, the state could use 9.5- or
or congestion. At the very least, this may reduce the need 10-foot lanes to “turn [a] four-lane express highway into a
for lane expansions. However, it is possible that AVs will en- six-lane express highway with literally the same right-of-
able road diets that accommodate the same amount of traffic way footprint” (McFarland 2015).
with fewer lanes. If AVs reduce the space required for vehicular traffic,
Traditional policies of roadway widening to support cities could retrofit the excess right-of-way to address a host
auto mobility may become irrelevant. According to Rich- of issues, including promoting active modes of transporta-
ard Biter, assistant secretary of the Florida Department of tion and stormwater management. The extra space no longer
Transportation, there may be opportunities to condense needed to move automobile traffic might be used to provide

Figure 4.1. The possible

transformation of a
typical neighborhood
streetscape in an AV
future (app.restreet.
com)

42 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

wider sidewalks, bike lanes, and more green space without re- and egress onto adjacent roadways. Consequently, the tran-
quiring more right-of-way. Take for example a typical urban sition from parking to drop-off areas will have far-reaching
neighborhood street section (Figure 4.1, top). Schlossberg, implications for access management, including the form, lo-
Riggs, Shay, and Millard-Ball (2018) suggest there is a strong cation, and design of curb cuts and drop-off/loading areas.
likelihood that on-street parking could be eliminated and AVs remove the need for passengers to be with the vehicle
lane width could be reduced to 10 or even nine feet on urban when it parks, enabling passengers to be dropped off instead
and suburban arterials. of having to exit the vehicle wherever parking is available. Us-
As illustrated on the bottom of Figure 4.1, this right- ers will likely want to be dropped off and picked up as close
of-way “recapture” is consistent with the idea of a road diet, to their destinations as possible. In this way, AVs will shift the
in which traffic lanes previously dedicated to automobiles priority at the site level from parking to drop-off areas. Drop-
are repurposed for other uses. AVs’ potential to reduce lane off areas will no longer be relegated to special uses like bus
widths could yield ample space that could be allocated for stops, train stations, and airports but will become a staple in
other purposes, perhaps even resulting in repurposed private the design of urban spaces.
driveways and garages, as well as public reuse of former on- Drop-off areas can take different shapes and sizes and
street parking areas (Schlossberg et al. 2018). could be incorporated into the designs of various urban set-
This also has implications for expanding complete tings in different ways. In many cases, the form of the drop-off
streets and bicyclist and pedestrian facilities. In this way, areas may be influenced by the existing built environment. In
AVs open the possibility of safer and more inviting urban fact, several features of today’s built environment could easily
places with more space for bicyclists, pedestrians, and other be retrofitted into drop-off areas, including turn lanes, front-
creative uses such as drop-off infrastructure (see the next age roads, and service roads. Since space will be at a premium
section). However, opportunities to recapture right-of-way in downtowns, there will likely be less space available for
to promote complete streets may be limited until AVs have downtown drop-offs, particularly separated drop-offs. How-
reached full adoption. Until every vehicle on the road is au- ever, if AVs shift the priority from parking to drop-offs, on-
tomated, lanes will still need to be designed to account for street parking spaces may be retrofitted into drop-off lanes.
human-driven vehicles. Although retrofitting existing infrastructure may pro-
Ultimately, AVs offer opportunities to create a built envi- vide opportunities for drop-off areas during the transition to
ronment that is more responsive to humans, but only if plan- AVs, once AVs are the predominant mode of transportation
ners and engineers are willing and able to prioritize moving drop-off/pick-up areas will likely be fully integrated into the
people over moving automobiles. With the emergence of design of almost all urban spaces. Drop-offs may take several
truly human-centered design—as opposed to automotive- different forms, including pull-offs, cul-de-sacs, and front-
centered design—there is also potential for ripple effects on age roads, but in all cases drop-off areas need to be separated
corridors and district land use. If AVs reinforce recent urban- from traffic lanes to ensure the safety of those entering and
ization trends, then rezoning or upzoning for superfluous exiting the vehicles. Although drop-offs and pick-ups are a
auto-serving uses, such as parking lots, service stations, and minor feature in today’s transportation system, they are ex-
repair facilities, may follow. At the same time communities pected to be one of the most important design elements in an
can and should plan for appropriate (potentially mixed use) AV-dominated world. Transportation engineers and planners
facilities that can service and charge high-tech vehicles. These will need creative ways to reuse existing infrastructure and to
land-use policies are discussed in greater depth in Chapter 5. develop completely new features to enable people to arrive at
and depart from their destinations safely and efficiently.

ACCESS MANAGEMENT
SIGNAGE AND SIGNALIZATION
The ability of AVs to drop off passengers before going to
park themselves or to pick up another passenger is expected Traffic signs and signals are among the most important fea-
to bring a drop-off revolution to the transportation system. tures of today’s transportation system. They provide drivers
Space previously used for on-site parking is expected to be with all the information they need to keep the transportation
transformed into drop-off areas, and businesses will likely system running smoothly and efficiently. Signs and signals
explore new site designs to allow for quick and easy ingress inform drivers when, where, and how fast they may go, and

www.planning.org AMERICAN PLANNING ASSOCIATION 43

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

ensure traffic keeps moving safely and efficiently through in- At the very least, this means that physical traffic signals
tersections. In short, traffic signs and signals are necessary to could be removed from intersections. However, the combi-
prevent the transportation system from devolving into chaos. nation of automated and connected vehicle technology may
However, the emergence of vehicle-to-vehicle (V2V) completely revolutionize how intersections function by re-
and vehicle-to-infrastructure (V2I) technology is poised to moving the need for traffic to stop at intersections. Instead,
revolutionize how information is transmitted to drivers and AVs able to sense and communicate with other vehicles will
how traffic moves through intersections. Although this tech- be able to flow freely through intersections. Each vehicle will
nology is not essential to AV technology, it will significantly simply react to other vehicles and cross traffic when an open-
improve traffic flow, reduce congestion, and help to bring all ing is available. While an overhaul of how intersections func-
of AVs’ potential benefits to fruition. The emergence of this tion (from a start-stop system to a free-flowing system) will
technology will also have important effects on the built en- likely not be possible until most or all vehicles on the road are
vironment. Most notably, automated and connected vehicles automated, these ideas promise to significantly improve traf-
offer the opportunity to declutter roadways by removing the fic flow and reduce congestion.
need for many traffic signs and signals.
The Reorientation of Signage and Signalization
The Digitization of Street Signs The only road signs and signals needed in an AV world are for
In the coming decades, information previously given to driv- pedestrians and bicyclists. AVs offer opportunities to reorient
ers through traffic signs (speed limit, road signs, stop signs, street signs and traffic signals from automobiles to pedestri-
etc.), will be transmitted to the vehicle through V2I sensors ans and redesign spaces to be more inviting to pedestrians
embedded in the roadway infrastructure. AVs can then adjust and other modes of transportation. Street signs could be re-
their speed, direction, or route according to the information placed by creative pedestrian wayfinding or other features
provided by the V2I “signs.” that make streetscapes more appealing and attractive to pe-
When AVs are first adopted and they share the road with destrians. In this way, AVs could make complete streets easier
human-operated vehicles, additional signs and signals may to implement, and could help create more attractive urban
be necessary to delineate where automated and human-oper- spaces and communities.
ated vehicles are and are not allowed to drive. However, once
AVs make up most or all of the vehicle fleet, physical signs Redesigning Roadways and Intersections
and signals are not expected to be necessary. Even lane strip- The geometry and capacity of existing roads is designed for hu-
ing could be phased out once vehicles can sense where other man drivers, so augmentations to accommodate AVs will be
vehicles are on the road. While some AVs currently rely on necessary—including how roads support multimodal travel
road lines to help their navigation systems, the primary func- for trains, bicycles, and pedestrians. As referenced in previous
tion of lane striping is to guide human eyes. More sophisti- sections, it is likely that neighborhood streets could be narrow-
cated systems may not need them, and they could be replaced er, intersections may not need signalization, and speed controls
by virtual lane systems embedded in the infrastructure. can become context sensitive (e.g., responding to inclement
In addition to serving as virtual signage, V2I technology weather or other conditions). Road diets may become easier,
could provide vehicles with real-time information on traffic given the increased level of service on existing roadways, and
delays and road work. AVs could then use this information to context-sensitive speed areas will be become more prevalent.
find the fastest and most efficient routes to their destination. Yet these road diets may be only a part of what plan-
While this could bring more traffic to neighborhood streets, ners need to anticipate. The design of roads will likely need
the transition from street signs to V2I technology could serve to evolve and city planners and policy makers will need to
to reduce traffic congestion and shorten travel times. be ready to work with transportation engineers to address
this. In a fully autonomous environment, reversible lanes or
The Decline of Traffic Signals advanced traffic control could be extended to dynamically al-
V2I and V2V technology will also contribute to the replace- locate major portions of infrastructure (Smolnicki and Sołtys
ment of traffic signals. Just like traffic signs, traffic signals will 2016; Bertozzi, Broggi, and Fascioli 2000).
no longer need to be visible; instead sensors embedded in the At the same time, the adoption of AVs may not support
road or placed in traffic towers will communicate traffic in- recent planning efforts to change one-way streets to two-way,
formation to vehicles on the road. increase bicycle and pedestrian levels of service, or create

44 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

more informal, Dutch woonerf-style streets that transform vestments in bicycle and pedestrian infrastructure it would
auto-dominated rights-of-way into shared spaces that accom- fundamentally remain an auto-oriented system. As AVs be-
modate all modes of transportation (Minneapolis Depart- come the dominant mode of transportation, travel by non-
ment of Public Works 2010; Brant 2016). While research has vehicular modes may be hampered by two key factors. First,
suggested one-way streets are less safe and efficient (Riggs and AVs require no signalization and signage to regulate traffic
Gilderbloom 2016a, 2017; Gayah and Daganzo 2012), driver- flow. As a consequence, red lights and stop signs that pro-
less cars may function better on one-way corridors than hu- vide for safe intersection crossings may become a thing of the
man-operated vehicles, and they perform better in roadway past. Pedestrians and bicyclists attempting to travel in dense
environments with formal rules and clearly indicated path- urban settings where traffic never stops could be left waiting
ways for different modes (Kelly et al. 2006). During the period for long periods for a break in traffic, slowing their travel.
of transition from few AVs on the road to AV predominance, Second, AVs will likely require regular drop-off and
AV technology will likely require mode separation. pick-up zones along most corridors. These zones allow riders
While advances in autonomous traffic control and re- in AVs to access their final destinations easily, as well as al-
versible lanes may provide opportunities to improve pedes- low riderless vehicles the ability to pick up passengers. These
trian facilities that planners should advocate for at the fed- zones require space for the AVs to access individual sites and
eral level, there is also a need to create protected streets that space for AVs waiting to pick up their riders. Depending
support bicycle and pedestrian travel and vibrant and attrac- upon their design and location, these loading and unloading
tive urban spaces and places. While this does not preclude zones could fragment bicycle and pedestrian networks and
shared streets or less formally designated pathways for vari- make travel via these modes more cumbersome. Poorly de-
ous modes, it does mean that rules of engagement and modal signed urban streetscapes that are dominated by AV drop-
priorities need to be clearly articulated. off and idling zones could have the effect of depressing bike
Riggs and Boswell (2016a) suggest that service expecta- and pedestrian travel.
tions should be codified for human-centered modes, some-
thing that will be further explored in the next section. This How AVs Might Support Bicycle
may provide a starting point in thinking about future road- and Pedestrian Travel
way infrastructure. Clearly AV technology will continue to While free-flow traffic conditions and drop-off zones may
evolve, and it is important that we co-evolve our thinking complicate bicyclist and pedestrian travel in urban areas, the
about street infrastructure to optimize safety in an autono- AV revolution also holds some promise for urban settings that
mous and connected system. serve humans first and vehicles second. AVs may require far
less space within urban settings; lane widths will become nar-
rower, fewer vehicles on the road will make road diets more
INTERFACE WITH BICYCLES AND PEDESTRIANS feasible, and the need to provide parking at every destination
will evaporate. Urban environments will also be less cluttered
The coming AV revolution offers substantial benefits for the by traffic signalization and signage, offering opportunities for
efficiency and safety of vehicular travel. Less clear, however, more attractive bicycle- and pedestrian-friendly corridors.
is the impact of AVs upon travel by bicyclists and pedestrians. Taken together, these possibilities suggest that road-
While AVs have the potential to improve the functioning of ways and urban environments could be redesigned in ways
vehicular systems, one view is that AVs may make bike and that will yield more enjoyable travel for bicyclists and pedes-
pedestrian travel within urban settings far more complicated trians. Many more urban corridors could become complete
and less easily achieved. Alternatively, because AVs will re- streets, with separate rights-of-way for AVs, bicyclists, and
quire less urban space than traditional vehicles, the technol- pedestrians. Reduced vehicular signage could open up op-
ogy offers some promise for the development of high-quality, portunities for signage and advertising aimed at bicyclists
attractive, separated bicycle and pedestrian infrastructure. and pedestrians, such as wayfinding signage, rather than
drivers and riders in vehicles. Surface parking and mono-
How AVs Might Hamper Bicycle lithic parking garages will become surplus and ultimately
and Pedestrian Travel can be replaced with more residential and nonresidential
The transition from human-driven to automated vehicles development or parks and plazas that serve as social spaces
promises to bring radical changes, but without significant in- and places for physical activity.

www.planning.org AMERICAN PLANNING ASSOCIATION 45

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

PARKING parking facilities for automated and human-driven vehicles

could capitalize on the space efficiencies of AVs while still
Without the need for a human driver to park the vehicle, the meeting the parking needs of human-driven vehicles. The
adoption of AVs will likely lead to a significant change in the most efficient parking paradigm to guide the transition to
design of parking in urbanized areas. As parking currently AVs would thus include on-site parking facilities designed
constitutes a significant percentage of the developed land in for human-driven vehicles and off-site parking facilities
urbanized areas, the impact of AVs on parking location, de- for AVs. Over time, the on-site facilities would gradually
sign, and amount required may be among the most signifi- shrink to accommodate the decreasing number of human-
cant changes to the built form of cities. Ultimately, far less driven vehicles on the road, while off-site AV parking facili-
space will need to be dedicated to parking, and the need for ties grow to take their place. As will be discussed further in
human-centered, human-scaled design for parking areas will Chapter 5, this likely would require cities to develop sepa-
be significantly reduced or eliminated once AVs become the rate parking standards for automated and human-driven
dominant mode of transportation. vehicles, including different dimensions and different mini-
mum parking requirements.
Parking Location Ultimately, parking structures for AVs will likely not
AVs are expected to significantly affect the location of park- need to take humans into consideration in the design and
ing facilities in urbanized areas. As AVs can drop passengers location of the facilities. Therefore, structured parking for
off at a destination and drive elsewhere to park, parking will AVs would not need to include human-friendly design and
not necessarily need to be provided on-site at every business, safety features such as lighting, elevators, and other ameni-
office, or residence. As a result, it is possible that parking in ties. Rather, AV structured parking could be located under-
urban areas could be consolidated outside of city centers. ground or in other underutilized or out-of-the-way loca-
Larger-scale, AV-only parking garages or lots could be located tions, with relatively little lighting or climate control. Some
on the periphery of urban centers where land values are not at parking in highly accessible areas may still be necessary to
a premium and development pressures are less intense. This allow for quick vehicle response when hailed, but in this
relocation of parking facilities away from the urban core will scenario, only limited access for humans would need to be
likely open up significant infill redevelopment opportunities. provided, such as maintenance stairways and other limited
Additionally, it is probable that on-street parking will facilities to allow for human access.
not be needed for AVs as human passengers will no longer
require parking at or near their destinations. The space once Reduction in Amount of Parking
used for on-street parking could be repurposed to accom- In addition to the substantial changes in where parking is
modate drop-off lanes or other right-of-way features, such as located and how it is designed, AVs are also expected to sig-
bike lanes or sidewalks. nificantly reduce the amount of parking required to meet
demand. AVs promise to do this in two ways: by making
Parking Facility Design Considerations more efficient use of existing parking and reducing the
The shift from on-site to remote parking facilities will not oc- overall demand for parking.
cur all at once, as human-driven vehicles will need to be ac- V2I communication could enable a more efficient use of
commodated on-site or nearby until the entire vehicle fleet is the parking supply by notifying AVs as to the location of the
automated. It is probable that during this transition, automat- nearest available parking spots. Previously underused park-
ed and human-driven vehicles may need separated parking ing would become an important part of the parking supply.
facilities to ensure AV efficiencies can be realized. For exam- More importantly, if AVs are owned under a shared
ple, AVs are expected to be able to park much closer together, model, some vehicles may not need to park after dropping
as vehicle doors will not need to be opened after the car parks off a passenger but would instead move on to the next hu-
itself. As a result, AVs could provide significant parking ef- man user. It is also possible that after passenger drop-off a
ficiencies in terms of the size and number of parking spaces privately-owned AV could return to the owner’s residence,
accommodated within a particular building footprint or sur- travel to a remote staging area for maintenance and fuel,
face parking lot. or circle the block instead of parking. All of these poten-
However, during the transition, human-driven vehicles tial factors would result in the reduction in the amount of
will still require parking designed for humans. Separate needed parking.

46 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

Figure 4.2. Conceptual site plan of

parking redevelopment opportunities
on the University of South Florida’s
campus (Chapin et al. 2016)

www.planning.org AMERICAN PLANNING ASSOCIATION 47

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

Figure 4.3. Example

of a typical
streetscape today
(Chapin et al. 2016)

Figure 4.4. Conceptual

design of a partially
automated streetscape
(Chapin et al. 2016)

48 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

Figure 4.5. Concep-

tual design of a fully au-
tonomous streetscape
(Chapin et al. 2016)

REDEVELOPMENT OPPORTUNITIES poses. Market demand may also lead to the redevelopment of
newly available land for residential or commercial uses.
By reducing the amount of needed parking and relocating A transformation in urban land uses due to AVs will un-
parking away from city centers, the widescale adoption of doubtedly occur gradually over a period of decades. During
AV technology will create significant urban redevelopment this transition period, urban settings where a clearer separa-
opportunities. In particular, the reductions and relocation tion between human drivers and AVs is possible could feature
of parking will allow the transformation of urbanized areas. prominent land-use changes. For example, enclosed environ-
Potential site design norms may be revolutionized with the ments where AV-only zones are more plausible, such as col-
ultimate adoption of a fully automated vehicular fleet. lege campuses, could be among the first places to experience
significant land-use changes. Figure 4.2 (p. 47) provides a
Transformation of Urban Centers vision for how parking at the University of South Florida’s
With the anticipated reduction in the number and size of Moffit Cancer Center could be redeveloped into research fa-
parking spaces, AVs provide an important opportunity to re- cilities, classrooms, and park space. Simply by redeveloping
think, revitalize, and redevelop urban centers. The reduction the surface parking (parking garages were left in place), the
and relocation of parking could open up significant land area university could increase the square footage of facilities by
for infill development within these areas. more than one-third and still have plenty of space left over
While market forces will play a large role in how newly for parks and green spaces. Land-use changes may also be
available land is redeveloped, policy makers will also have roles more prominent along highways with dedicated AV lanes,
in determining how publicly owned land once used for wider along AV-only drop-off and pick-up areas, and in the areas
roadways and parking lots will be repurposed. Underutilized surrounding AV-only parking facilities.
parking lots could become an important resource that could
be used to achieve community goals. Available land could be Site Design
used to enhance placemaking or beautification efforts, or for As on-site parking is reduced, site plans for commercial
recreational facilities, parks, drainage, or other similar pur- and residential development will change. Local develop-

www.planning.org AMERICAN PLANNING ASSOCIATION 49

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

Figure 4.6. AVs’ transfor-

mation of a city block—
before and after (Chapin
et al. 2016)

ment regulations will likely reduce—or completely elimi- ing. In areas with high traffic volumes, multiple drop-off ar-
nate—on-site parking requirements. This will result in an eas could be designed to mitigate crowding and congestion.
evolution of site planning from a focus on meeting parking Commercial strip centers alongside major highways could
requirements to designing building envelopes and ameni- consolidate entrances and exits, increasing buildable area
ties that more efficiently use sites. on a site-by-site basis and potentially improving traffic flow.
Since AVs will likely drop off passengers at their desti- This could also allow for fewer turn lanes and the ability to
nations, newly constructed buildings may be located closer repurpose existing turn lanes. Access roads off main thor-
to the abutting road frontage and site plans may include oughfares may also be redesigned to act as drop-off areas
drop-off areas that are adjacent or connected to the build- easily accessible from highways.

50 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 4

NEW DESIGN PARADIGMS environment in our communities. In particular, the wide-

spread use of AVs could have major impacts on roadways
By combining all six of these major types of built environ- and rights-of-way, access management, signage and signal-
ment changes, we can begin to get a picture of what the future ization, pedestrian and bicycle facilities, parking, and de-
in an AV world might look like. This section provides two velopment opportunities. However, none of these land-use
visualizations of how AV technology may transform public changes will happen in a vacuum independent of the other
spaces over time. impacts. Right-of-way implications will affect pedestrian
Figures 4.3 through 4.5 (pp. 48–49) show how a typical and bicycle spaces. Revolutionary changes to intersections
arterial road might change over the next 50 years. The first and signage will change both the experience of those in vehi-
image (Figure 4.3) depicts a roadway in Tallahassee, Florida, cles and that of pedestrians and bicyclists near the roadway.
in 2016. It is characterized by wide lanes, frequent curb cuts, Parking implications will affect demand for drop-offs and
large traffic signals, and parking located in front of buildings. opportunities for redevelopment. Site design will change to
As AVs begin to be adopted (Figure 4.4), some notable address this new demand.
changes are expected, including separated AV lanes. These In short, by coupling a denser urban environment with
lanes are narrower and allow room for added bicycle lanes. the ability to narrow rights-of-way and enhance bicycle and
Turn lanes, medians, and traffic signals remain. pedestrian facilities, AVs provide planners and policy makers
Once AVs become the dominant mode of transportation with excellent opportunities to revitalize urban centers and
(Figure 4.5), all of the lanes could become narrow AV lanes, create thriving urban spaces. However, the technology will
which allows for separated or protected bike lanes and more also create major challenges to capitalizing on these oppor-
attractive pedestrian infrastructure. In other areas, drop- tunities and achieving these goals. Careful planning will be
offs would replace on-street parking or be prioritized over required to shape AVs’ built environment impacts to create
bike lanes. Medians would disappear, and in some cases lane safer, more efficient, and more vibrant communities. The fol-
striping itself may be a thing of the past. Traffic signals would lowing chapter will provide planners with information and
be replaced with V2I communication infrastructure. Pedes- guidance on how to begin proactively planning for the com-
trian wayfinding signage will be the only signage needed. ing built environment impacts.
Instead of typical strip mall-style buildings with parking in
front, buildings can move up to the street. Overall, what was
an uninviting arterial whose primary purpose was simply to
move traffic as fast as possible has now been turned into an
attractive, people-friendly, efficient, and safe urban environ-
ment that could draw people in and provide more inviting
spaces to live, work, and play.
Similarly, Figure 4.6 (p. 50) shows how AVs could trans-
form a typical city block. Before the adoption of AVs, most
city blocks needed to include a significant amount of park-
ing to accommodate the block’s residents and commercial
patrons. This included large surface lots as well as on-street
parking. However, in an AV world, on-street parking could
easily be retrofitted into a drop-off lane providing easy ac-
cess to the blocks’ destinations. Since vehicles could drop off
passengers instead of parking, surface lots become significant
opportunities for urban infill development.

CONCLUSION

Like many technological advances before it, the rise of AV

technology is poised to have a significant impact on the built

www.planning.org AMERICAN PLANNING ASSOCIATION 51

CHAPTER 5
CONSIDERATIONS
FOR POLICY
MAKING AND
INFRASTRUCTURE
INVESTMENTS
As described in Chapter 4, autonomous vehicles will cause the next great transformation, not only in our transportation sys-
tems, but in the entire built environment. How planners respond to this opportunity will shape the impact AVs will have on
our communities.

By proactively leveraging AV technology to create ADDRESSING PLANNING OPPORTUNITIES

great urban places, planners could reshape urban areas in AND CHALLENGES
ways that promote safe, sustainable, and people-centered
environments. However, a concerted effort will be re- Like any major technological innovation in transportation,
quired if planners are to take advantage of AVs’ placemak- AVs offer a host of important opportunities and difficult chal-
ing potential. lenges that planners will have to navigate if they are to work
The challenges described in Chapter 3 suggest that with- toward building better communities. These issues were in-
out careful planning, AVs may: (1) exacerbate auto-oriented troduced in Chapter 3, and here we tie back directly to these
sprawl, (2) increase vehicle miles traveled (VMT) and emis- challenges, as addressing them appropriately leads to much
sions, (3) adversely affect active modes of travel and public of the promise of AVs. It is vital for planners to understand
health, (4) impact transit, and (5) lead to more acute social the implications of AV technology for their communities and
equity issues. to develop policy and infrastructure solutions that will ad-
In a white paper published by the American Planning dress them. To that end, this section is intended to provide
Association, Henaghan et al. (2018) present four principles to guidance on how to capitalize on AVs’ opportunities while
guide AV planning efforts. These principles are: mitigating their challenges.

• The time to begin planning is now. Addressing Impacts on Auto-Oriented Sprawl

• Good planning principles still hold. As discussed in Chapter 3, it is expected that AVs will both
• Planning must anticipate the disruptive effects of technol- promote sprawl and support reurbanization at the same time.
ogy, both positive and negative. By reducing the cost of travel and enabling people to work
• Planning must account for uncertainty. or relax during their commute, AVs threaten to exacerbate
sprawling development patterns by allowing people to live
Planners need to begin proactively applying sound even further away from urban centers. However, AVs will
planning principles and best practices to the comprehen- simultaneously provide opportunities for reurbanization by
sive impacts of AV technology, as outlined in Chapters 3 opening underutilized parking lots for infill development
and 4, in ways that consider a range of possible futures and and by supporting shared mobility services that would bol-
are flexible enough to account for rapid changes in tech- ster urban, carless lifestyles. Planners’ policy responses to
nology and behavior. To do this, planning agencies may this challenge may affect whether AVs ultimately encourage
need to move toward more iterative, scenario-based plan- sprawl or more compact development.
ning processes to inform their long-range planning efforts.
This chapter will provide planning policy ideas and infra- Discouraging Auto-Oriented Sprawl
structure investments that apply these principles first to If AVs reduce travel time and stress so that living farther from
AVs’ opportunities and challenges and then to AVs’ built city centers is not a burden, the only factors keeping devel-
environment impacts. opment from moving farther into exurban and rural areas

www.planning.org AMERICAN PLANNING ASSOCIATION 53

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

would be land-use regulations, the relative cost of develop- Local jurisdictions should identify appropriate areas to
ment, and demand for an urban lifestyle. Land-use regula- locate AV parking, and determine how peripheral parking
tions that constrain exurban development and promote in- structures may influence adjacent land uses. Special consid-
fill may be the primary ways to prevent further sprawl and eration may need to be given about whether to locate these
encourage urban revitalization. However, since AVs open the parking structures in industrial areas, or whether other ar-
potential for long commutes that extend well beyond juris- eas on the urban fringe are more appropriate. These areas
dictional boundaries, AVs will also elevate the need for re- should be appropriately designated in comprehensive plans
gional growth management strategies that consistently apply and other planning documents. Also, once specified areas
growth management tools across jurisdictional lines. Region- are designated for these parking facilities, transportation
al strategies would require extensive coordination efforts, but agencies will need to consider whether appropriate road in-
they could work to constrain development to activity centers frastructure exists to serve these parking locations or if ad-
within each jurisdiction. ditional roadway capacity will be necessary. If the majority
To enhance these growth management efforts, munici- of AVs are electric vehicles, then parking facilities will also
palities could also consider providing incentives to redevelop need to be designed to incorporate charging facilities. Fi-
underutilized parking facilities into higher and better uses. nally, as noted above, local governments may consider if and
The reduction in the demand for parking opens new oppor- how they might incentivize the redevelopment of existing
tunities to create vibrant urban spaces through infill devel- parking facilities to revitalize urban cores and make better
opment. However, given the relative expense of redeveloping use of largely underused parking spaces.
parking facilities compared to greenfields, these incentives State, regional, and local agencies must also consider
may have to be coupled with growth management strategies how to revise building and engineering codes to accommo-
(e.g., urban growth areas) that focus development. Otherwise date the design and construction of consolidated AV parking
the incentives are not likely to be strong enough to entice in- structures. Building codes may need to be revisited to relax
fill development and to prevent greenfield development from requirements for human-centered amenities in parking fa-
stretching farther and farther away from city centers. cilities, such as climate control, passageways, turning radii,
If shared AVs develop into a primary mode of trans- elevators, and potentially even lighting.
portation, shared mobility service may naturally reinforce
growth management strategies. Similar to transit systems Addressing Impacts on Increased
today, shared mobility providers could likely provide better VMT and Vehicle Emissions
service within dense urban environments. If AVs, especially As noted in Chapter 3, the lowering of both perceived and
shared AVs, encourage a decline in car ownership as some monetary costs of travel by AV has the potential to increase
researchers have suggested (Fagnant, Kockelman, and Bansal automobile travel by commuters, riders that have tradition-
2015; McDowell 2014; Schoettle and Sivak 2015), many peo- ally relied on transit or other transportation services, and
ple may be dependent on shared service and would be more nondrivers. VMT and the resulting vehicle emissions also
likely to live near urban centers for the mobility benefits. In have the potential to increase through “zero-occupancy”
this way, shared AVs may naturally mitigate sprawl and en- trips as empty AVs travel between use and parking. An im-
courage compact development. portant factor in both reducing the overall number of AVs on
the road and achieving maximum efficiencies in their use is
Rethinking Parking the shared use of AVs.
Parking is one of the key links to the land-use and sprawl-
related impacts of AVs. AV technology is expected to trans- Promoting Use of Shared AVs
form the design and location of parking infrastructure in As was introduced at the beginning of this report, many have
our urban centers. As the need and demand for on-site speculated that a large part of the AV future will be shared. In
parking is significantly reduced, and ultimately eliminated, fact, many researchers have predicted that the three revolu-
parking infrastructure is likely to be pushed to and con- tions in urban transportation will be the automation, elec-
solidated at the city periphery where land is more readily trification, and sharing of the transportation system (Fulton,
available and affordable. For policy makers, regulatory con- Mason, and Meroux 2017). While this has been largely fu-
siderations include issues related to the permitting of con- eled by the rise in transportation network companies (TNCs)
solidated and structured parking. such as Uber and Lyft, and by the urban lifestyles of many

54 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

millennials, it also is a product of vehicle cost. Many manu- access to downtowns, business districts, or academic and
facturers are pursuing driverless electric fleets as they roll out medical campuses, especially in the most congested areas.
their highly autonomous (Levels 4 and 5) vehicles. This shift Such restrictions can be accomplished through street clo-
to shared mobility will be key in capitalizing on the planning sures, traffic calming, and time-of-day pricing or access.
opportunities and addressing the planning challenges identi- • Look at data sharing and behavior. At the same time
fied in Chapter 3. Shared AVs could reduce VMT compared as they are managing and encouraging shared use of AV
to single- and zero-occupancy vehicles and could also pro- platforms, planners and policy makers need to explore op-
mote social equity by reducing the cost of travel. portunities and partnerships to assess and facilitate shar-
Highlighting the land-use and environmental benefits ing behavior. Efforts to create partnerships with TNCs are
of car- and ride-sharing services will be an important part especially important as they capture travel data that would
of planners’ roles going forward, particularly since research be useful for planning purposes. At the same time, it is im-
is already showing that these services are likely to result in portant to develop transportation demand management
increased reliance on on-demand access to cars (Clewlow and policies, including monetary and behavioral strategies,
Mishra 2017). Planners need to consider multiple policies, that reduce the number and duration of trips.
particularly in the arena of equitable access, as the highest
levels of vehicular access will continue to be in urban cen- Reevaluating Access Management
ters with higher levels of density to support these services. As An important piece of the shared-AV puzzle is providing ap-
such, this could increase the gap between the urban rich and propriate infrastructure. As AVs grow in popularity, more
the poor (both rural and urban), and planners should work to and more users will likely want to be dropped off near their
implement the following policies. destination instead of parking farther away and walking.
Designated areas for dropping off and picking up AV passen-
• Strengthen and link to TDM efforts. Policies should be gers will likely become a common feature for roadways and
linked to transportation demand management efforts and site plans. However, if unregulated, these areas could cause
discourage increased and longer trips. Some partnerships congestion problems by backing up traffic into streets.
are already being explored by TNCs to reduce single-oc- To accommodate drop-off areas without creating traffic
cupancy vehicle ownership and use (e.g., Lyft’s pilot trip or safety problems, state and local agencies will need to create
reduction program). Incentives for vehicle sharing might new design standards for drop-off areas and drop-off lanes.
be established and linked to development conditions of ap- In particular, specifications for the length, number of drop-
proval. Further shared mobility policy frameworks can be off points, or number of lanes required for a site’s drop-off
found in PAS Report 583, Planning for Shared Mobility (Co- area may be necessary. These standards need to be tailored to
hen and Shaheen 2017). Table 5.1 (p.56) provides a summa- the size and expected demand of drop-off areas to prevent a
ry of three policy frameworks outlined in PAS Report 583. traffic backup. Separate standards for different categories of
• Plan for increased densification and intensification. drop-off areas may even be necessary (e.g., shopping malls
Similar to transit systems, shared mobility and TNCs would require larger drop-off areas than small businesses).
are more viable in dense urban environments as higher In high-density areas such as downtowns, it may also be im-
densities of riders improve the efficiency and cost-effec- portant to separate drop-off areas from pick-up areas to make
tiveness of these systems. Continuing to pursue sound the arrival and departure process as efficient as possible and
planning principles and strategies to create vibrant urban to ensure that drop-off areas do not impede the flow of traffic.
environments by progressively increasing the density and Another option to reduce the number of ingress and egress
intensity of development will support the use of shared points in dense urban areas would be to strategically place
AVs and improve the cost-effectiveness of shared mobil- drop-off areas to serve an entire block (Dennis et al. 2017).
ity. These initiatives will need to take different forms in Regardless of the setting or the design of drop-off/pick-
different contexts, as appropriate densification strategies up areas, another vital feature of pick-up areas in particular
should be tailored to downtowns, other urban districts, will be passenger waiting areas. Riders will need safe and
and suburban areas. comfortable spaces to wait for their vehicles. The design of
• Explore car-free downtowns. In conjunction with in- these areas could borrow cues from existing best practices
creasing the prevalence of drop-off zones (discussed fur- of bus stop design, including the need for shaded and cov-
ther below), consider limiting single-occupant automobile ered places to sit.

www.planning.org AMERICAN PLANNING ASSOCIATION 55

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

TABLE 5.1. SHARED MOBILITY POLICY APPROACHES FOR LOCAL GOVERNMENTS

Shared Mobility as an Shared Mobility as a Shared Mobility

Environmental Benefit Sustainable Business as a Business
(maximum governmental support) (moderate governmental support) (minimum governmental support)

Jurisdiction may allocate public rights-of- Jurisdiction maintains a highly formal-

way (such as parking, loading zones, etc.) Jurisdiction that once allocated public ized and established process for the allo-
Allocation of
on a case-by-case basis or through more rights-of-way through an informal pro- cation of public rights-of-way, including
Rights-of-Way
informal processes, such as nonbinding cess formalizes this process. a process for allocation among multiple
council/board of director resolutions. operators.

Fees may be based on cost recovery of

providing rights-of-way (e.g., fees based
Recognizing the social and environ- Fees are based on a cost-recovery or
on foregone meter revenue and admin-
mental benefits of shared mobility, profit-based methodology. This could
istrative costs) associated with providing
Fees and Permits jurisdiction provides public rights-of-way include permit costs, lost meter revenue,
on-street parking. In other instances, fees
free of charge or significantly below and administrative expenses for program
may be reduced to reflect environmental
market cost. management.
goals, such as charging a reduced car-
pooling rate for car-sharing parking.

Jurisdiction pays for the sign installation Jurisdiction pays for the installation, and Jurisdiction requires shared operators to
Signage, Markings,
and maintenance, striping, and markings the operator pays for the maintenance of pay for the installation and maintenance
and Installation
associated with the shared modes. signage, striping, and markings. of signage, striping, and markings.

Jurisdiction requires that shared opera- Jurisdiction may require that shared mo-
Social and Jurisdiction does not require any social
tors study and document local social bility operators study and document lo-
Environmental and environmental impact studies of
and environmental impacts at regular cal social and environmental impacts on
Impact Studies shared mobility.
intervals. a one-time basis or at regular intervals.

Informal process, if any, led by the

jurisdiction to elicit public input into the Informal process where the jurisdic- Highly formalized process where shared
location and scaling of shared modes on tion and shared mobility operator seek mobility operators are responsible for
Public and
public rights-of-way. For example, staff public input into the locations of shared obtaining public input and approval
Stakeholder
may internally determine the location services through public notification and on the locations of services through
Involvement
and number of car-sharing parking staff management of possible public neighborhood councils, commissions, or
spaces or public bike-sharing stations concerns. formal hearings.
without public comment.

Source: Cohen and Shaheen 2017

Local governments may also consider making some drop-off areas could consistently be integrated into small
changes to existing zoning codes to better accommodate area and corridor plans to ensure sufficient vehicle access is
drop-off areas, including reducing the required number of provided while minimizing conflicts with bicyclists and pe-
parking spaces, reducing setbacks, and specifying drop-off destrians. Specific policies are outlined as follows:
design standards. In effect, parking requirements may be
replaced by specifications for the number of drop-off points • Establish locations for pick-up and drop-off zones for
required for a site’s drop-off area depending on the number passengers and deliveries. To ensure lively downtown
of trips the destination was projected to generate. In addition, environments, cities should put curb policy into practice

56 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

by first considering appropriate locations for drop-offs. pecially in the near term. Much of this involves rethinking
Such locations should be areas with high curbside activ- the right-of-way in terms of modal priorities, which academ-
ity (existing or planned) for passengers and goods. Cities ics and practitioners are beginning to do.
should determine where to best locate drop-off zones to This section is not intended as a step-by-step guide to
serve several buildings at a time and to minimize con- navigate the transition to an AV world, but rather as a starting
flicts among modes. This could include use of side streets, point for preparing for AVs’ influence on the built environ-
alleys, or hubs. For example, in 2017, New York City ad- ment. Key considerations to incorporate into infrastructure
opted a suite of measures to ease roadway and curbside investments and redevelopment decisions are highlighted as
congestion. This initiative involves six city departments they relate to different elements of the built environment.
and comprises five focus areas, including curb lane flow
and curb access restrictions (New York 2017). Such initia- Rethinking and Reprioritizing Rights-of-Way
tives offer a preview of how to manage the growing de- Retrofitting existing roadways and rights-of-way to accom-
mand for curbside access. modate the efficiencies of AVs may be among the first and
• Design pick-up and drop-off zones to minimize con- most important steps for state and local transportation agen-
flicts and optimize flows. Design is important to mini- cies to prepare for the emergence of AVs. Agencies will need
mize conflicts, facilitate seamless transfers, and moderate to adjust how streetscapes are designed and engineered, as
flows of multiple vehicles. For example, many buildings well as how and where roadways are planned and built. When
use circular driveways for passenger pick-up; however, this changing roadways to accommodate new vehicle types, plan-
creates two curb cuts, increases stress for pedestrians, and ners and policy makers should highlight the benefits of de-
occupies valuable sidewalk space. It also creates a complex signs that use placemaking to retrofit roadways and create
navigation system for autonomous vehicles of multiple strong and healthy places.
types: cars, transit vehicles, and delivery ground drones.
Rethinking the designs of pick-up and drop-off zones will What will not change:
be important to improve flows and minimize conflicts in • The placemaking imperative and how walk/bike integra-
the AV future. tion is key to gaining multiple benefits, economic benefits
• Create pick-up and drop-off regulations and manage- in particular
ment schemes. Controlling the operation of vehicles on • The importance of efficiently moving more people through
the street will be essential to manage the constant flow space
of vehicles expected with shared autonomous travel and • The need to manage multiple modes of transportation
deliveries. Again, New York City’s new measures to man- • The need for variety in street networks and designs to
age congestion may provide a glimpse of how to prioritize serve a variety of contexts and users
and manage flows among personal, transit, and delivery
vehicles, as well as pedestrians and other sidewalk uses. What will change:
Operating parameters can be programmed to limit or • The consideration of rights-of-way as amenities rather
provide access; for example, AVs can be limited to certain than infrastructure to move cars
locations in the city or their curb access limited to cer- • The ability to digitally control traffic (in addition to con-
tain hours. Curbside regulation should prioritize transit to trolling via curbs, medians, paint, and signage)
ensure the rise of shared AVs does not prevent transit ve- • The need for redundant systems, given technology’s vul-
hicles from accessing the curb at bus stops (NACTO 2017). nerabilities
• The introduction of speed as a determinant of street man-
agement (for example, there will be low, medium, and
Addressing Impacts on high-speed automated vehicles; there is already growing
Active Travel and Public Health tension between electric bikes or scooters and human-
The impacts of AVs on active travel and public health could powered bikes)
be quite significant, particularly if they make automobile use
easier and induce more demand. That said, there are numer- Schlossberg et al. (2018) recommend three basic princi-
ous opportunities to reinforce planning values and shape the ples in rethinking streets: thin the lanes, remove parking, and
built environment to support AVs alongside other modes, es- think shared. These provide general guidance for policy de-

www.planning.org AMERICAN PLANNING ASSOCIATION 57

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

velopment and are also consistent with what planners might Obispo, California (Table 5.2). The planning document al-
consider “good planning.” They might serve as primary goals, locates funding along the same lines.
but what specifically should planners and policy makers be • Integrate AVs into current plans. Proactively address-
doing to rethink their roads? This is an important consider- ing AVs in current plans is vital because the planning
ation that revolves around the two basic approaches to right- horizons of most long-range transportation plans extend
of-way design—planning and engineering. well beyond when AVs are projected to become avail-
First, in the planning sphere, it is important that AV con- able. At the regional and local levels, agencies will need
siderations be integrated with the design of streetscapes and to begin considering the impact of AVs on travel demand
road networks. Trip generation, level-of-service (LOS) models, and throughput as they develop their long-range plans.
and planning documents for all modes need to take AVs into In particular, regional transportation planning bodies
account. In that light the following policies are of great import. will need to reconsider whether future lane-expansion
projects will be necessary if AVs reduce congestion and
• Create a modal hierarchy for roadway space and modes. increase the throughput of each lane. To inform this pro-
Local plans should create modal hierarchies that frame cess, continuing research assessing the effect of AVs on
project and funding priorities. These hierarchies can be throughput and travel demand will be necessary to pro-
broken out by roadway type (e.g., neighborhood, collec- vide guidance on how many lanes are required and where
tor, arterial) or be consistent citywide. They should set road diets could be appropriate.
policy thresholds for how roadway space can and should • Transition to VMT-based models. Planning documents
be allocated in planning and transportation documents. should reflect the potential for AVs to reshape roadways
An example of this kind of modal hierarchy is provided in and increase efficiency, and they should plan for that
the circulation element from the general plan of San Luis space. In some cases that will mean accepting a LOS that is
below local thresholds, as well as moving to using models
based on VMT or person-miles traveled instead of tradi-
tional LOS models. VMT models are already being used
TABLE 5.2. CITY OF SAN LUIS OBISPO, CALIFORNIA, in many cities in the U.S. These models plan for trips and
MODAL HIERARCHY trip length while not penalizing dense urban development
through the environmental review process by allocating
Complete Streets Areas Priority Mode Ranking them greater numbers of auto trips than more suburban
developments with fewer people. Emerging policy must
1. Pedestrians
focus on VMT as an environmental impact (regardless of
2. Bicycles
Downtown & Upper Monterey Street
3. Transit
fuel type or efficiency standard) and underscore efforts to
4. Vehicles shift away from LOS traffic analysis.
• Expand LOS analyses to include pedestrian, bicycle, and
1. Pedestrians transit service. LOS analyses have traditionally focused
2. Bicycles solely on service to automobiles. Expanding LOS analyses
Residential Corridors & Neighborhoods
3. Vehicles to account for other modes of transportation, particularly
4. Transit
at the local level, will help to shift the focus from moving
1. Vehicles vehicles to improving the mobility for everyone. This will
2. Bicycles also serve to ensure investments in AV infrastructure do
Commercial Corridors & Areas
3. Transit not fragment bicyclist and pedestrian networks or rein-
4. Pedestrians force an auto-oriented transportation system, as was dis-
1. Vehicles
cussed in Chapter 3.
2. Transit • Incorporate AVs into transportation demand models.
Regional Arterial & Highway Corridors AVs will need to be incorporated into travel demand mod-
3. Bicycles
4. Pedestrians eling standards and practices. Pilot projects and ongoing
testing of AVs, particularly AVs in real-world settings, will
Source: Table 3.3, City of San Luis Obispo General Plan, Circulation Element be a vital part of informing the data and assumptions un-
(2017) derpinning the demand modeling process. Once AVs are

58 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

Figure 5.1. Right-of-way

recapture concept
(Appleyard and Riggs
2017)

Existing Single-Family Housing Stock/Lot

Recaptured Right-of-Way (Civic or Private Benefit)

New Multifamily Recaptured Right-of-Way

integrated into modeling methods, they can be an integral given similar or equal investment in comparison to au-
part of informing the long-range transportation planning tonomous vehicle infrastructure and street re-design.
process determining when and where new roads, roadway
expansions, and potential lane reductions are necessary. In this context, Appleyard and Riggs argue that larg-
As technology improves, planners could even begin ex- er lanes may be reduced and the space left over from the
ploring how the same algorithmic and machine-learning narrower lanes (indicated in blue in Figure 5.1) should be
processes that run AV technology might inform new ways apportioned back to bike lanes, pedestrian malls and side-
of thinking about activity-based transportation models in walks, play areas, and even housing through right-of-way
land-use and transportation planning processes. recapture programs. They suggest that this could be a value
proposition for cities and for residents because streets could
Secondly, in the engineering realm, roadway design transition to other uses, such as housing, on the recaptured
manuals need to be refreshed so planners and engineers can right-of-way. On larger rights-of-way there could even be
take advantage of opportunities for multimodal and complete opportunities for creative small-unit multifamily dwellings
street solutions that harness the power of potential right-of- (indicated in orange in Figure 5.1). Given the opportunities
way gains. As suggested in a 2018 Transportation Research this creates, cities might consider deeding back this real es-
Board paper by Appleyard and Riggs (2017): tate to private owners to increase municipal property tax
revenue on an annual basis.
In neighborhood right-of-ways and areas of high pe- While this warrants more dialogue among the planning
destrian activity, pedestrians, cyclists and vulnerable and research community, it clearly underscores the oppor-
road users should be protected by (a) reduced auto tunity that AVs offer in terms of reusing and repurposing
right-of-way; (b) routing traffic onto appropriate streets rights-of-way, which can serve to benefit mobility and pub-
(commercial over residential) and recognizing road lic health at the same time through increasing walking and
hierarchy; and (c) by limiting speed of AVs to 20 kilo- biking and mitigating the potential negative impacts of AVs.
meters per hour to improve safety and livability…Pe- Planning organizations should look to the following policies
destrian and cycling based infrastructure [should] be when rethinking the design and engineering of roadways.

www.planning.org AMERICAN PLANNING ASSOCIATION 59

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

• Adopt bicycle- and pedestrian-friendly roadway design ter of street may dictate the opportunity this presents.
standards. Local planners should build upon complete For example, Schlossberg et al. (2018) show that a typical
streets best practices. They should consider design for 80-feet-wide, four-lane arterial with two lanes of parking
bicyclists and pedestrians from the beginning of any AV offers an opportunity to remove the parking, thin and
system-related planning initiatives and infrastructure de- reduce the lanes, and provide space for 40 or more feet
signs and avoid impairing the pedestrian experience. for bikes, pedestrians, and transit. In the spirit of experi-
• Transition additional roadway capacity to bicycles, mentation happening in places like Barcelona, where of-
pedestrians, transit, or shared vehicles. Rights-of-way ficials are shutting down every other street to auto traffic,
should be reused in ways consistent with modal priori- cities may decide to eliminate roadways altogether and to
ties—most notably to support modes that are environ- use those spaces for parks, housing, or other needed uses.
mentally friendly and promote public health. Excess As shown in Figure 5.2 (p. 61), ultimately policies and
rights-of-way should be used for facilities such as dedi- plans should enable streets to be transformed to achieve
cated bike or transit lanes and expanded, more attractive community goals—for example, affordable housing and
sidewalks, as opposed to increasing auto mobility. This linear parks. Excess right-of-way provides a significant
is consistent with the June 2016 National Association of opportunity to green our communities and to creatively
City Transportation Officials policy statement, in which manage stormwater.
the first priority of planning for AVs is that they “promote
safety for pedestrians, bicyclists, transit riders, automated Balancing AV Needs with Bicycles and Pedestrians
vehicle passengers, and all street users within the multi- Paralleling the importance of prioritizing roadways for ac-
modal urban context” (NACTO 2016). tive travel, one key challenge during the AV revolution will
• Rethink smart road and intelligent transportation sys- be balancing the needs of AVs with the needs of other travel
tems roadway infrastructure and invest in basic road- modes. This balance is particularly important for bicycle and
way infrastructure. Within the automotive industry pedestrian travel in urban settings—trips that offer physical,
there has been much debate over the necessity of “smart” mental, and emotional health benefits, as well as environ-
infrastructure, given the costs. Literature indicates that mental benefits due to the use of human effort rather than
fully autonomous vehicles will eventually be capable of fossil fuels or electric power. However, AVs may lead to a pre-
handling all urban and rural driving conditions without dominance of free-flow intersections and will require pick-
surrounding roadway infrastructure. Given this, it might up/drop-off areas at more locations, which make travel better
be wise to pause all but the most basic technology infra- for AV riders—but at a cost to bike and pedestrian travel.
structure investments—particularly vehicle-to-every- If transportation and land-use planning agencies are
thing (both infrastructure and vehicles) communication proactive, AVs can be integrated into urban settings in ways
platforms. This technology may play an important role in that encourage rather than hinder travel by bicyclists and pe-
the future, but it is still evolving. To best optimize funds, destrians. To do so requires attention to these non-AV travel
planners should pursue consistent design standards for modes from the beginning of any AV-related system plan-
signage and lane markings across all roadways; federal, ning initiatives and infrastructure redesigns. Assessing and
state, and local roads should not have different standards. then factoring in the needs of bicyclists and pedestrians at the
Especially in the short term, it may be more cost-effective outset promises greater efficiency in the transition and less
to pursue basic infrastructure investments. This could al- expense in the long run.
low technology to advance using consistent infrastructure In anticipation of the transformation of streetscapes,
that driverless machine learning processes can detect (e.g., transportation planners and engineers need to adapt road-
signage, lane markings, etc.). The more consistent policy way design guidelines to factor in how AV technology will
and standards are across space, the easier it will be for AVs impact these systems. Building upon best practices in the
to achieve full functionality. complete streets literature offers a path forward, although
• Explore creative and experimental right-of-way use. these guidelines will need to evolve to take into account the
Planners should explore new uses and the innovative different functionalities of AVs. As noted above, guidelines
ways streets can be used, thinking beyond the typi- will also need to be developed for the frequency, design, and
cal streetscape. There are many ways that a traditional size of AV pick-up and drop-off areas to ensure that these ele-
street might change, and the existing size and charac- ments do not fragment bicycle and pedestrian networks.

60 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

Figure 5.2. A post-AV

street might become a
linear park (Schlossberg
et al. 2018)

Important system or infrastructure changes will be re- and interacts with these modes. While Google claims that its
quired to promote bicyclist and pedestrian mobility in an AV AVs will be able to detect and share the road safely with bicy-
world. Solutions will be needed to allow for safe and regu- clists (Rosen 2016), and there have been experiments at test
lar crossing of free-flow intersections in busy urban settings. sites around the country to resolve some of these challenges
These solutions can take the form of regularized, dedicated (Messner 2017; Marshall 2017), there is still much work to be
bicycle and pedestrian crossing periods, priority signaliza- done in advancing the technology.
tion for pedestrians, or separated systems for crossing busy In light of that, planning academics Riggs and Boswell
streets in the form of tunnels or bridges. The overriding issue (2016c) have called for standards to advance certain levels of
is the need to ensure that bicyclist and pedestrian mobility safety, particularly for cyclists. These call for a baseline ser-
is not compromised by the transition to an AV-only system. vice standard to guide technological development and the re-
The greatest promise for successfully accommodating lated policy decisions that regulate the relationship between
AVs along with bicyclists and pedestrians in urban settings bicycles and AVs. As a starting point, these principles provide
revolves around the extra available space within the right-of- a series of expectations to guide both policy and technology
way that will follow from the narrower vehicle lane widths. development between active transportation modes and AVs
This surplus pavement holds opportunities for dedicated bike (see sidebar on p. 62). Local governments can support this
lanes and expanded, more active sidewalks, spaces that are approach by adopting the following planning policies to sup-
no longer as cluttered with signage and signalization aimed port bicycle and pedestrian travel.
at vehicular traffic. Given that these surplus spaces should
be systemwide, there will likely be opportunities for trans- • Embrace recommended practices in bicycle and pedes-
forming existing roadways into dedicated bike boulevards, as trian planning. Cities should establish recommended
found in many European cities. practices in planning and design for bicycles and pedes-
In this way, AVs present an opportunity to rethink trians. This includes integrating multimodal LOS analysis
streets, and particularly to support bicycle and pedestrian in infrastructure design and planning, adopting “vision-
travel. While there are design opportunities, there are also zero” plans and policies that aim to eliminate traffic fa-
potential concerns with regard to how AV technology detects talities, and supporting NACTO guidelines that provide

www.planning.org AMERICAN PLANNING ASSOCIATION 61

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

BICYCLING MANIFESTO FOR AN AV FUTURE

Since bicyclists will share the road with 11. AVs should minimize travel on streets
AVs, they have the greatest chance of designated as bicycle boulevards or
conflict. Riggs and Boswell (2016c) devel- that have high bicycle usage but no
oped 13 principles as a manifesto for cy- facilities.
clists in an AV world. These are intended 12. Companies deploying shared AVs
to provide a “starting point for dialogue” should ensure adequate supply of ve-
to promote the safety of cyclists sharing hicles equipped with bicycle racks or
the road with AVs. carriers to meet demand.
13. AV companies should record and
1. AVs should be able to detect bicy- share all collision data with local, state,
clists and detect and understand all and national law enforcement and
bicycle signage and lane markings. regulatory agencies.
2. AVs should be able to detect and un-
derstand bicyclists’ hand signals.
3. AVs should cede the right-of-way to
bicyclists.
4. AVs should have an ability to signal
(visual and audible) their detection of
bicyclists and pedestrians and basic
intent.
5. AVs should follow bicyclists at a safe
distance when unable to pass.
6. AVs should exceed the three-foot
minimum passing rule, especially as
speed increases.
7. AVs should leave an ample margin of
safety when making decisions about
turning, passing, ceding right-of-way,
and other decision-making scenarios
involving bicyclists.
8. AVs should be able to detect ap-
proaching bicyclists and prevent
“dooring” (instances when cyclists
are struck by the doors of parked cars
along a roadway as passengers exit).
9. AVs should be designed (size, shape,
weight, materials) to minimize injury
to bicyclists should an impact occur.
10. AVs should travel at speeds appropri-
ate for urban conditions to facilitate
safe travel for nonautomotive users
(for example, not more than 20 miles
per hour on downtown and neigh-
borhood streets, 40 miles per hour
on arterial connectors, etc.).

62 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

high LOS to nonautomotive users. Further, cities should that the mobility needs of transit-dependent populations
consider adopting more aggressive policy-based service are being met. If AV technology increases the cost of own-
standards for cyclists and pedestrians consistent with the ing and operating a personal vehicle, it will be especially
expectations set by Riggs and Boswell. important to provide adequate public transit service to ac-
• Advance design in transportation engineering. While commodate the travel needs of those who may not be able to
engineered solutions often fit within a finely described box afford an AV. Integrating AV technology into public transit
of what is acceptable under the terms of state and federal service may be the best way to ensure AVs do not exacerbate
manuals, namely the U.S. Federal Highway Administra- transportation equity issues.
tion’s Manual on Uniform Traffic Control Devices, this can As discussed in Chapter 3, AV technology will provide
be an impediment to innovation at the local level. Plan- transit agencies with significant opportunities to improve
ners should encourage and educate local engineers on rec- transit service by improving safety and reducing operating
ommended practices, and should help them experiment costs. To begin capitalizing on the benefits that AVs could
with new interventions to provide innovative yet afford- provide, transit planners must pilot transit applications of AV
able ways to enable pedestrians and cyclists to engage with technology. Numerous jurisdictions, such as Las Vegas, are
and cross through AV traffic and intersections. piloting or implementing fully autonomous transit routes on
• Continue to invest in and develop transportation de- public roads across the country.
mand management (TDM) and other travel behavior Universities are also leading the way in autonomous
programs. Many scenarios on the future of transporta- transit adoption, as the University of Santa Clara began test-
tion predict changes in consumer and housing prefer- ing an autonomous shuttle in November 2016. Other univer-
ences; and while those are shaped by individual behaviors, sities, including the University of Michigan, the University
they are also shaped by the policy and programmatic en- of Florida, and the University of Cincinnati, are preparing
vironment. Planners should continue to design programs to introduce autonomous shuttle service in 2018. The lessons
to support activity and health-centric behaviors, includ- learned from these early adopters should pave the way for
ing walking, bicycling, and social interactions. Existing other jurisdictions to implement autonomous transit service.
outreach and TDM programs should be enhanced so that In the early stages of AV adoption, retrofitting exist-
they continue to offer balanced transportation choices ing Bus Rapid Transit (BRT) routes with autonomous buses
that clearly articulate the benefits of walking and cycling presents an outstanding opportunity to incorporate AV tech-
to enhance and sustain active transportation (Aitken et al. nology into the transit system. This is especially true of BRT
2016), but also contribute to other factors that have been routes with dedicated infrastructure that would enable the
found to frame travel choices—including socialization vehicle to operate on a dedicated lane without the need to ac-
and understanding of environmental impacts (Jariyasu- quire more right-of-way or to reduce the number of vehicle
nant et al. 2015; Riggs 2017). New programs might include lanes. Providing AVs with dedicated infrastructure reduces
walking and bicycling lunch programs, promotion of af- the number of variables the autonomous technology must ac-
ter-dinner strolls at restaurants, or ride sharing, particu- count for and prevents the vehicle from needing to execute
larly for people who have yet to establish work trip habits. more difficult driving maneuvers, such as changing lanes in
heavy traffic. Dedicated infrastructure would also serve to
Addressing Impacts on Public ease public concerns over the safety of driving with AVs.
Transit and Other Roadway Uses Another way dedicated bus lanes can support autono-
Along with ensuring that the needs of bicyclists and pedestri- mous vehicle navigation is by having consistent and infor-
ans are addressed when transitioning to an AV future, planners mative lane markings that will help guide the bus. Magnetic
must also consider the potential impacts of AV on public transit markers could help the vehicle navigate in weather condi-
and other uses of the roadway system, including freight. tions that make Lidar less accurate. In fact, researchers have
already designed and deployed an automated steering control
Integrating AVs into Transit Systems system made up of magnetic markers on an EmX BRT Line in
Autonomous and shared technology has the potential to Eugene, Oregon (Huang and Tan 2016). This control system
dramatically influence the way we think about transit. It is consists of magnetic markers under the roadway every me-
vital to make sure the provision of quality transit service ter or so to provide information to the bus for lane-keeping
does not become a low priority in an AV world to ensure assistance and precision docking that allows the bus to pull

www.planning.org AMERICAN PLANNING ASSOCIATION 63

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

right up to the edge of the curb at each station. On rainy or

TABLE 5.3. AV CANDIDATE PERFORMANCE snowy days, the bus is still able to navigate seamlessly because
MEASURES AND FRAMEWORKS of the magnetic markers helping to guide it with location in-
formation. Similar systems with increasingly higher levels of
TRANSPORT PERFORMANCE MEASURES automation are likely to soon follow. Mercedes-Benz has pro-
Transport Effectiveness grammed its new Future Bus (an autonomous bus) to operate
• Percent population served in bus-only lanes to make the design easier to develop and
• Revenue passengers per service area population implement (Thompson 2017). It could also provide an excel-
• Total passengers per vehicle lent opportunity to pilot CV signal prioritization technology.
• Revenue passengers per revenue vehicle mile
In short, retrofitting BRT lanes with AV technology pro-
• Revenue passengers per revenue vehicle hour
vides an excellent opportunity to test autonomous buses in
Street Level Multimodal/Livability Effectiveness real-world environments to determine whether they can pro-
• Percent vehicle sharing vide the promised cost savings in a way that is safer, more
• Speeds
feasible, and more publicly acceptable than simply replacing
• Vehicle size
• Percent nonautomotive buses on traditional bus routes. Yet retrofitting BRT routes is
• Miles of roads over 25 MPH just one example of the types of autonomous pilot programs
• Intersection density that could help to jumpstart the transition to autonomous
Transport Efficiency transit. The lessons learned from these early tests will be vital
• Revenue vehicle miles per vehicle to promote the use of AV technology.
• Total vehicle miles per vehicle For many years public transit has struggled with finan-
• Revenue vehicle hours per vehicle cial pressures as well as service, ridership, and reliability is-
• Operating expense per seat mile sues. A second transit-related change will be the ability to
• Operating expense per revenue vehicle customize transit vehicles and vehicle types to fill different
• Mile operating expense per total vehicle mile
roles than in the past. This can be used to eliminate some
• Operating expense per revenue vehicle hour
• Energy consumption per revenue vehicle mile
of the challenges of proximity, specifically the first- and last-
• Energy consumption per total vehicle mile mile issues that relate to the initial and final portions of a
• Energy consumption per revenue vehicle hour transit-based trip between home or work and a destination
• Operating expense per total passengers (Scott 2017). New forms of transit will require thinking about
• Operating expense per revenue passenger investments and innovations in current systems.
• Operating expense per passenger mile Autonomous last-mile solutions are sometimes referred
LAND-USE PERFORMANCE MEASURES to as micro-transit, but as Larco has suggested (2017), these
Land-Use Accessibility Effectiveness new forms of transit raise challenges in terms of service and
• Jobs / housing balance integration. While the rise of shared AVs is often discussed
• Land-use mix (entropy) as creating a future that eliminates the need for fixed-route
• Number mixed use developments / housing units transit, he argues that:
• Human scale / intersection density
Land-Use Efficiency Both Lyft and Uber have recently made moves to make
• Sprawl indices their ride services more efficient by having riders walk
• Regional land consumption to higher volume streets. Lyft has introduced “Pickup
• Greenbelt acres
Suggestions” and Uber has their “walk to the corner
DEHUMANIZATION INDICATORS system”—incentivizing people to walk to the nearest
• Large automated vehicles avenue or arterial instead of being picked up on a more
• Fast automated vehicles minor street. This reduces travel time which allows
• Pedestrian and bicycle intimidation more people to be picked up (without becoming frus-
• Perception of safety for biking and walking
trated as they wait).
• Sociocognitive and physiological health
• Superhuman-scale streets, blocks, development
From a city development perspective, this points to
Source: Appleyard and Riggs 2017 the continued importance of higher volume streets as

64 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

Figure 5.3. Local delivery

robot pilot in Redwood
City, California (William
Riggs)

transit hubs (even if it is not traditional transit) and Integrating Automation into Freight and Delivery Service
begs the question of how to make pickups and drop- Along with transit operations, the movement of goods is an
offs most efficient along these high-volume routes (a important consideration for cities. Freight is expected to be
designated spot on each block?). There are obvious one of the first sectors to adopt AVs. Automated technology
benefits to having some kind of hierarchy in micro- has been an important part of the freight industry for de-
transit—how should we design streets that can ac- cades. Robots transport containers from ships to railcars,
commodate this? automated freight trains transport goods across the country,
and highly automated distribution centers have significantly
While we have discussed the design implications for improved the efficiency of freight logistics. The rise of AVs
streets in a previous section, the notion of how we assign will simply extend freight automation to semi-trucks and
transit access to these corridors is an important one—and delivery vehicles.
one that will need focus as our systems become reinvented. The primary reason the trucking industry is excited
Planners will need to focus on reinventing roadway access about autonomous technology is the cost savings the tech-
through seamless fare and payment integration. They will nology will provide. In addition to potentially reducing labor
also need to consider performance standards, an area that costs, AV technology could reduce fuel cost by more than
has been explored by Appleyard and Riggs (2017). 10 percent through the aerodynamic improvements of pla-
In considering AV performance measures, Appleyard tooning (ITF 2017). Since trucking fleets turn over twice as
and Riggs have suggested building on the work of Fielding quickly as consumer vehicles, autonomous trucks likely will
to develop candidate measures and metrics of performance penetrate the market significantly faster than consumer cars
(Fielding, Glauthier, and Lave 1978; Fielding, Babitsky, and once they become available.
Brenner 1985). They provide a conceptual framework and list While the automation of freight and delivery services
of candidate performance measures that could be useful in may have narrower impacts on urban form and on the built
helping to evaluate the livability, efficiency, and functionality environment, it is vital to carefully consider the ramifications
of a transportation system that includes AVs (Table 5.3, p. 64). autonomous technology will have upon freight. Specifically,

www.planning.org AMERICAN PLANNING ASSOCIATION 65

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

this technology will have notable implications for traffic op- Addressing Impacts on Social Equity
erations, traffic safety, travel behavior and demand, and the As was discussed in Chapter 3, the social equity implications
infrastructure necessary to support freight. Planning for au- of AVs are highly significant and echoed throughout this doc-
tonomous freight will also involve special considerations as its ument. While this issue touches on many of the items already
development will extend beyond the automobile to unmanned discussed with regard to land-use considerations and transit
aerial vehicles or drones. access, it is important for planners to encourage and lead dia-
One of the biggest implications of autonomous trucking logue around the development of more specific policy—par-
for transportation planning is the need to address platoon- ticularly in the areas of accessibility and jobs.
ing. Although connecting “trains” of autonomous trucks In the final chapter of his book, Disruptive Transport
may provide significant efficiency benefits, during the transi- (forthcoming), William Riggs discusses the magnitude of
tion period when AVs and human-driven vehicles must share these potential issues:
the road, platooning could create safety problems for human-
driven vehicles trying to navigate around truck platoons. …what happens when transit and shared rides become
One option to address this issue would be to dedicate lanes owned by a private company. Should that company
for freight usage, but this likely would require significant in- have the right to deny you access to their platform?
frastructure investments. A more cost-effective option may Should they have a right to deny cities data about the
be to restrict the length of platoons to two to four trucks. This vehicles they are running on city roads? Should they
would enable the efficiency benefits of platooning while also have a right to leave a city, or a portion of a city without
ensuring proper safety standards. notice? This could yield a trouble reality of data-driven
Consistent with trends in the consolidation of parking at transportation, meaning that certain parts of the city,
the city periphery, there will likely need to be a transition for perhaps those that are less dense, less safe or less profit-
land at the periphery of cities to adjust to new distribution- able, become less served by future mobility. Lyft CEO
and logistics-oriented uses. Facilities that move goods from John Zimmer recently pledged transportation equity
long-distance trucks to local delivery vehicles will likely also and to provide service to low-income communities as
be consolidated at the edges of urban areas. Planners and they roll out scooters (Zimmer and Green 2018). Yet,
policy makers should consider flexible zoning overlays along- absent policy, what is the long-term certainty of such
side greenbelt protections in these areas to facilitate this use, a commitment. There may always the temptation to
while limiting potential environmental impacts. capitalize on or profit from (as opposed to serve) popu-
At the local level, freight should be integrated and seam- lations that are under-represented and have limited
less and allowed to evolve innovatively. Autonomous delivery transportation access.
systems are already being tested across the world (Edelstein
2018; Marakby 2018; Nagata 2018). In the U.S., Amazon and Ensuring that the benefits of AVs and disruptive trans-
Google have begun parcel delivery and 7-Eleven has been portation are accessible is just one challenge, and he goes on
piloting delivering Slurpees via drones (Hidalgo 2017). Non- to discuss another issue of equal importance: jobs.
automotive logistics (e-bike, cargo bike, or handheld parcel
delivery) could reduce traffic from multiple sources ranging With replacement of professional human drivers there
from cars to drones to delivery robots at the neighborhood is a need to address how these jobs will be replaced in
level (Figure 5.3, p. 65), thereby promoting the livability of the a fair and equitable manner. Job loss and replacement
local communities. will not be limited to ridesharing jobs but result in far
Expanding on this idea, planners will need to explore reaching impacts that affect everyone from traditional
how new and emerging forms of freight impact users at the taxi drivers, mail carriers, and freight and cargo driv-
local level. Scenarios are already being painted of a world in ers, to name a few.
which there is constant movement of autonomous vehicles,
but it will be important to ensure this does not have adverse Estimates vary about the ultimate impact of these two is-
effects on neighborhoods or quality of life. Partnership and sues, access and jobs, but the research is clear: They will likely
experimentation with shared autonomous technology com- have a disproportionate impact on vulnerable and at-risk
panies will be important for achieving the benefits of this populations that have historically been ignored as a part of
technology while mitigating concerns related to it. planning and land-use processes. There are actions that cit-

66 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

ies can take now to buffer potential negative impacts of new to pricing and infrastructure burdens. Acceptable policies
transportation innovations. These include potential policies should be linked to transportation demand management
for local governments outlined below—again, this list is not efforts and not create regressive effects that inhibit mobil-
intended as a prescriptive roadmap but an agenda for dia- ity for low-income residents or certain geographically con-
logue and customization in every community. centrated areas. Fleet service companies should be able to
demonstrate standards that ensure AV access, perhaps by
• Maintain robust transit service and explore becom- sharing scrubbed data showing access throughout a city.
ing an AV provider. With rising budgets and new mo- This is particularly important for shared AVs.
bility options, the temptation for cities can be to reduce • Protect vulnerable roadway users. Underscoring the call
public service in favor of mobility-on-demand or paid- to protect cyclists and pedestrians in the prior policy pro-
service/for-profit options. Cities should set clear priori- posals, cities have a social equity impetus to adopt vision-
ties in comprehensive and general planning documents zero plans and adopt vision-zero guidelines to protect us-
to provide transit access, particularly in locations with ers who may have no auto access.
less density where cheaper, private-sector service may not
be available. They should also conduct feasibility analy-
ses and explore the viability of becoming autonomous TRANSITIONING TO AN AV-DOMINATED SYSTEM
transit providers. There are many emerging autonomous
shuttle providers that can function well in dense, urban, Many of the major changes to the built environment and the
and even mixed traffic scenarios (Figure 5.4). These may benefits AV technology could provide will not be viable until
be able to better serve at-risk populations at lower cost most or all of the vehicle fleet is made up of AVs. Roadways
than existing transit platforms. and parking spaces shared by autonomous and human-driv-
• Establish access standards for TNCs and fleets. Policy en vehicles will still need to be designed to safely accommo-
makers could set up policies to maintain equitable access date the latter. For example, narrower pavement widths will
to public spaces and infrastructure and to enforce adequate not be possible until virtually 100 percent of the vehicle fleet
privacy standards. This should include shared mobility is autonomous, because human-driven vehicles may not be
policies relating to social equity, particularly with regard able to safely navigate narrower streets. However, the transi-

Figure 5.4. Cities might

partner with autono-
mous shuttle providers
to maintain efficient
service in certain loca-
tions (May Mobility)

www.planning.org AMERICAN PLANNING ASSOCIATION 67

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

tion period in which human-driven and autonomous vehicles falling asleep or drifting into oncoming traffic. The on-street
will share the road is not expected to be short. As discussed parking has also been removed from one side of the street to
in Chapter 2, it typically takes about 15 years to replace the accommodate the gradual decline in the demand for parking.
majority of a vehicle fleet (Kuhr et al. 2017). While the novelty All of these reductions create space to add a bike lane to both
of AVs may help to speed this process, the higher price tag of sides of the street.
autonomous technology may prolong the transition. As AVs grow in popularity and approach full adoption,
Consequently, before planners will be able to pursue more significant right-of-way changes may become pos-
many of the important opportunities described in this re- sible. The third image in Figure 5.5 shows one example of
port, careful consideration will need to be given to facilitat- a fully autonomous right-of-way. Assuming all vehicles are
ing a smooth transition to an AV fleet. One option would autonomous, lane widths for all four lanes can be reduced
be to simply wait until AVs are the predominant mode of to eight feet, drop-off lanes can replace on-street parking on
transportation before implementing the policy changes and both sides of the street, and all of the additional space can be
infrastructure investments outlined in the previous sections. transformed into a protected bike lane and wider sidewalks.
However, passively awaiting full adoption would repeat the With the same total right-of-way, AVs can transform the
same mistake planners made during the rise of the automo- streetscape, making room for bicycle and pedestrian facili-
bile by allowing the technology, rather than good planning ties without causing more congestion. As discussed in Chap-
principles, to shape development patterns. ter 4, full automation could open the door to more intensive
Proactive planning will be required throughout the tran- streetscape improvements, such as road diets. The fourth im-
sition to ensure that AV technology is used to create urban age provides an example of this.
spaces that are safer and more efficient than ever before. In Providing dedicated lanes for AVs would allow for two
addition, proactively planning for AVs during the transition lanes that are smaller and more efficient than traditional ve-
could enable communities to enjoy some of the safety and ef- hicle lanes. This could improve congestion and provide more
ficiency benefits the technology promises to provide during space for bicycle and pedestrian facilities years before AVs
the transition instead of waiting for full adoption. reach full adoption. In this way, the second image in Figure
5.5 provides an example of the types of roads that could be
Rights-of-Way in Transition constructed during the transition to an AV-dominated sys-
A prime example of these benefits can be seen in how roads tem. Making these infrastructure improvements while AVs
and rights-of-way could undergo iterative transformation are being adopted would be vital for enabling a faster and
processes to capitalize on the efficiency of AVs without com- smoother transition to fully autonomous roadways. In Fig-
promising the safety of human-driven vehicles. ure 5.5, apart from the protective barriers for the bike lanes,
The top image in Figure 5.5 (p. 69) shows a simple repre- the only changes that would be required to move from the 50
sentation of a typical right-of-way today. With four 11- to 12- percent AV right-of-way to the 100 percent AV right-of-way
foot lanes, some on-street parking, a median, no bike lanes, would be to restripe the lanes. This would prevent the need
and a relatively narrow five-foot sidewalk, this auto-oriented for massive infrastructure investments when AVs approach
road is primarily designed to move traffic as quickly as pos- 100 percent adoption. Investing in dedicated AV infrastruc-
sible. As discussed in Chapter 3, AVs’ ability to travel close ture throughout the transition period would help to spread
together could increase throughput and traffic flow. However, out the investments necessary to capitalize on the benefits
during the transition from human-driven vehicles to AVs, AVs could provide.
platooning AVs could distress or endanger human drivers Consequently, an important consideration during the
unless separated from human-driven traffic. early stages of AV adoption will be identifying and prioritiz-
With these considerations, capitalizing on AVs’ efficien- ing roadways where dedicated AV lanes could be implement-
cy benefits without compromising the safety of human-driv- ed. State and federal highways may present easier opportu-
en vehicles during the transition will likely require separate nities for dedicated lanes initially because they have simpler
infrastructure for autonomous and human-driven vehicles. traffic patterns, fewer intersections, and fewer points of in-
The second image in Figure 5.5 provides an example of a gress and egress than local roadways. However, over time
roadway with dedicated AV lanes. The inside AV lanes are many local roads could likely benefit from AV-only lanes.
several feet narrower than traditional driving lanes. The me- Although dedicated AV lanes present the need for system
dian has also been removed because AVs are not in danger of prioritization, future transportation systems may also ben-

68 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

Figure 5.5. Potential

streetscape transforma-
100 Percent Human Driven
tions as AVs are adopted
100 Percent Human Driven
100 Percent Human Driven (app.restreet.com)
100 Percent Human Driven

Parking Drive Lane Drive Lane Drive Lane Drive Lane Parking
Lane Lane app.restreet.com
Parking Drive Lane Drive Lane Drive Lane Drive Lane Parking
Lane Lane app.restreet.com
Parking Drive Lane Drive Lane Drive Lane Drive Lane Parking
Lane
Parking Drive Lane Drive Lane Drive Lane Drive Lane Lane
Parking app.restreet.com
Lane Lane app.restreet.com
50 Percent Human Driven; 50 Percent AV Driven
50 Percent Human Driven; 50 Percent AV Driven
50 Percent Human Driven; 50 Percent AV Driven
50 Percent Human Driven; 50 Percent AV Driven

Parking Drive Lane Drive Lane Drive Lane Drive Lane

Lane app.restreet.com
Parking Drive Lane Drive Lane Drive Lane Drive Lane
Lane app.restreet.com
Parking Drive Lane Drive Lane Drive Lane Drive Lane
Lane app.restreet.com
Parking Drive Lane Drive Lane Drive Lane Drive Lane
Lane app.restreet.com
100 Percent AV Driven
100 Percent AV Driven
100 Percent AV Driven
100 Percent AV Driven

Dropoff Drive Lane Drive Lane Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
100 Percent AV Driven Alternative
100 Percent AV Driven Alternative
100 Percent AV Driven Alternative
100 Percent AV Driven Alternative

Dropoff Drive Lane Drive Lane Dropoff

Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com
Dropoff Drive Lane Drive Lane Dropoff
Lane Lane app.restreet.com

www.planning.org AMERICAN PLANNING ASSOCIATION 69

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

efit from similar prioritization systems for the transition of Planners can use visioning to start a community con-
intersections, parking structures, and virtual infrastructure. versation about AVs, educating the public about the potential
impacts of AVs and the importance of preparing with pro-
Transitioning to AV Parking active policy making. In Florida, a team from Florida State
As seen in Figures 4.3–4.5 in Chapter 4 (pp. 48–49), while University engaged in award-winning visioning sessions
right-of-way changes enabled by the transition to AV use are about the future of AVs. They asked participants to envision
taking place, new opportunities for infill development are ex- how AVs would impact their communities and helped them
pected to arise as parking is able to be redeveloped into high- develop a variety of design solutions to help achieve their vi-
er and better uses. This will provide opportunities to catalyze sion of the future. For example, participants placed a strong
vibrant urban spaces that are compact and walkable. emphasis on providing separate facilities for automated and
During the transition to AVs, as discussed in Chap- human-driven vehicles during the transition to AVs. These
ter 3, planners should support on-site facilities designed for findings helped to expand the AV planning discussion in the
human-driven vehicles as well as off-site parking facilities state of Florida to include considerations of this technology’s
to capitalize on AVs’ smaller parking footprints. Over time, implications for the built environment and community de-
the area devoted to on-site facilities would gradually shrink sign (Chapin et al. 2016). The findings also helped to inform
as the numbers of human-driven vehicles decline. Off-site many of the built environment impacts discussed in Chap-
AV parking facilities would be developed to take their place. ter 4 of this report. While these sessions primarily engaged
Gradually transitioning to off-site parking would open the planning practitioners for research purposes, similar vision-
door to more efficient land-use patterns. ing sessions could be conducted with local communities to
A reduction in the demand for parking would also of- identify how they would like to integrate AVs and other smart
fer opportunities to begin replacing on-street parking with technologies into their communities. Planners and policy
pick-up and drop-off areas. As discussed earlier in this chap- makers can then use this information to backcast with spe-
ter, local planners may want to establish guidelines for the cific actions (both design and policy) that communities could
placement of these initial drop-off areas. Land-use codes take to achieve those visions.
could provide developers with the option of providing either Likewise, this same process of visioning can be inter-
a certain number of parking spots or a pick-up/drop-off area, twined with scenario planning and alternatives generation.
or require them to replace a certain number of parking spots Since AVs could have a wide range of impacts, visioning ef-
with a pick-up/drop-off area. forts often produce several different visions of the future.
Scenario planning processes can help develop and plan for
each of these visions. This could offer a more robust approach
ADDRESSING AV IN COMMUNITY to preparing for AVs by providing the flexibility to prepare
PLANNING PROCESSES for and adapt to multiple visions of the future. For example,
in the realm of transit, Milam and Riggs (2018) discussed
Beyond understanding the potential impacts of AVs on the looking at a range of options for planning and developing
built environment and the challenges and opportunities this alternative scenarios based on the following principles: in-
technology will bring to communities, planners will also creasing frequency of service; extending operational hours;
need to consider how the methods of planning practice might providing transit-only lanes; automating transit service; bet-
need to change to address AV technology. This report focuses ter matching or “right sizing” transit demand to type of ser-
on the built environment implications of and policy consid- vice; and integrating equity into service performance. Sce-
erations for AVs, but planners should also be thinking about narios can be modeled in different ways and then mapped
how to integrate this new technology not just into modeling to distinct policy outcomes. For example, this could include
but into planning processes and community engagement. establishing vehicle requirements (clean energy, shared use,
Two strategies, visioning and alternatives generation, etc.) implementing roadway fees, making changes in land
will be especially important in addressing AVs within com- use, or extending a greenbelt.
munity planning processes. These strategies offer the oppor- Given the uncertainty surrounding AVs, scenario plan-
tunity for citizens and policy makers to look at a range of fu- ning processes that prepare for a range of options are better
ture alternatives and choose the vision and policy outcomes positioned to develop plans agile enough to adapt to changes
that best suit their community goals. in technology and travel behavior. In addition, focusing on

70 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 5

incremental actions can help create plans capable of adapting ship. Regional growth management and land-use plans need
to future risk and uncertainty. Policy actions can be mapped to be regularly reevaluated to stay ahead of the ways AVs influ-
to actions that are triggered when certain thresholds or situa- ence travel behavior and exurbanization trends. Consistently
tions arise. While this type of nimble action may require con- updating comprehensive plans and other long-range planning
stant revisiting through annual reporting and staff-level data documents every few years is vital to ensuring that plans ad-
analysis, it is essential as AVs shape our cities in ways we only just to the disruptive nature of AVs (Henaghan et al. 2018).
partially understand at this point in time. Balancing the need to be nimble with the need to be pro-
active will be especially challenging in the area of infrastruc-
ture investments, as their lifespans are long and it is not cheap
CONCLUSION or easy to retrofit roads or bridges. However, as seen in the
example of utilizing dedicated AV lanes to gradually transi-
The planning policy response to AV technology should be tion to AV roadways while human-driven vehicles are still on
comprehensive in scope, addressing a range of issues includ- the road, planners may need to adopt a paradigm of gradual,
ing streetscape design standards, parking requirements, bi- iterative infrastructure investments to remain nimble enough
cycle and pedestrian facilities, curbside management, traffic to adjust to changes in the technology and how it is used.
and intersection operations, and local and regional growth In spite of, or better yet, because of the radical changes
management strategies. Yet policy efforts will need to be tar- that AVs are expected to bring, it is vital that planners promote
geted to strategically address key concerns at the right time in sound planning principles and placemaking practices that en-
order to capitalize on opportunities within a rapidly evolving sure access for all, including the most vulnerable groups. The
field. Planners cannot wait for full AV adoption to begin pre- desire to create attractive, vibrant places where people want
paring for the technology’s effects. Proactive planning policy to live, work, and play has not changed. The planning profes-
will be necessary to avoid making the same mistakes made sion has long struggled to balance the demands of growing the
during the rise of the automobile. economy, protecting the environment, and supporting equity
However, given the uncertainty surrounding the de- for all, and these challenges will not abate with the coming of
velopment, adoption, and usage of AVs, planners must also the AV age. While technology can improve our lives, there is
embrace an approach of continuous learning. The technology no guarantee that it will improve the lives of all people.
is evolving so rapidly that today’s best practices may not be This chapter has attempted to present some of the ways
relevant in five to ten years. Planners need to stay abreast of planning practice will need to evolve and adapt as AVs be-
technological innovation, federal and state regulation, and come an integral part of the transportation system, but the
the implications of early adopter successes and failures to in- overarching goals and many of the best practices in planning
form long-range planning efforts. practice and urban design will not change. Consequently,
Planning processes may also need to be refined to better promoting the use of AV technology at the expense of other
account for the uncertainty surrounding the future in an AV planning goals would be counterproductive. It is vital that
world. Proper community planning responses to AVs will be planners do not allow the novelty of AV technology to dis-
in part dependent on the technology’s deployment timeline tract them from striving to create livable, sustainable, and
and predominant ownership model (private versus shared). affordable communities. Instead, planners need to find in-
Planning efforts may need to consider a range of possible fu- novative ways of utilizing AV technology in pursuit of these
tures to account for this uncertainty. Consequently, scenario established planning goals. This report hopes to provide a
planning and visioning need to become regular parts of long- starting place for these efforts.
range planning efforts to address the range of possible futures.
In addition to addressing this uncertainty, planning pol-
icy for AVs will need to be nimble and flexible to adapt to the
technology’s rapidly evolving capabilities. Planning interven-
tions should be phased in over the long transition period from
human-driven to autonomous vehicles and should be adjust-
able in order to react to new developments in technology and
behavior. For example, parking requirements need to be read-
justed over time to account for declines in automobile owner-

www.planning.org AMERICAN PLANNING ASSOCIATION 71

CHAPTER 6
CONCLUSION: THE
TRANSFORMATIVE
IMPACTS OF AVS
Autonomous vehicle technology promises to reshape the transportation system and the built environment in ways not seen
since the introduction of Henry Ford’s Model T more than a century ago. By revolutionizing the nature of personal mobil-
ity and removing the need for passengers to be in the car at all times, AVs have the potential to dramatically impact roadway
design and the built environment to yield urban spaces that are safer, more efficient, and more attractive.

There is compelling evidence that AVs will allow for • Promote community conversations that identify risks and
smaller and more efficient rights-of-way, increase the speed benefits associated with AVs, set goals and priorities, and
and throughput of roadways, and open up spaces for bicy- identify potential policy and funding levers to better meet
cle and pedestrian facilities, green spaces, and other urban goals and harness benefits.
amenities. The ability of AVs to wirelessly obtain information • Work now to incorporate AV considerations into the de-
on destinations, traffic patterns, and intersections promises sign of streetscapes and road networks, including revisit-
to declutter urban environments by removing traffic signs ing roadway design manuals and long-range transporta-
and signals. Drop-off and pick-up areas are expected to re- tion plans.
place parking lots as the predominant locations for passenger • Begin identifying ways to establish separated AV infra-
entry and exit. As the majority of parking is relocated into structure, such as dedicated AV lanes, to ensure that the
consolidated parking facilities away from urban centers, efficiencies of AVs can be capitalized on during the early
large amounts of previously underutilized space will be made stages of AV adoption.
available for redevelopment opportunities. • Convene design and development stakeholders to develop
Much like the Model T of the early 1900s, AVs will usher standards for the size and location of drop-off areas to ac-
in massive changes in the way people travel, the form and commodate the growing demand for drop-offs without
function of our transportation systems, and the look and feel backing up traffic.
of the environments in which we live, work, and play. How- • Investigate opportunities to use excess pavement and
ever, unlike the American experience with the Model T, it is rights-of-way to promote complete streets through pro-
hoped that this time policy makers will recognize and take tected bicycle and pedestrian infrastructure, active
advantage of this opportunity to reshape our urban areas in streetscapes, and green spaces.
ways that promote safe, sustainable, and people-centered en- • Task researchers and engineers with developing innova-
vironments. AV technology offers an opportunity to balance tive yet affordable ways to enable pedestrians to safely
what have long been seen as conflicting goals of safer and cross free-flowing AV intersections.
more efficient transportation systems on the one hand and • Recognize and begin planning for changes in parking
urban environments founded upon the principles of sustain- demand by identifying long-term opportunities for AV
ability and human-centered design on the other. parking structures or large surface lots away from city
The twin goals of efficiency and urbanity can be achieved centers, revising codes for parking requirements, and in-
only through proactive planning and investment by federal, corporating AV parking areas into comprehensive plans
state, regional, and local transportation agencies. As intro- and other planning documents.
duced in this report, agencies need to look to the following
concepts to best capitalize on the opportunities afforded by Clearly, further research is required to assess how best to
AVs to create safe, efficient, and livable places: integrate AVs into the transportation system and to under-
stand how AVs will reshape the built environment. This re-

www.planning.org AMERICAN PLANNING ASSOCIATION 73

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 6

port provides a first step toward envisioning the future in an

AV world, a future that can yield attractive, people-friendly,
efficient, and safe urban environments. To achieve that fu-
ture, transportation and land-use planning agencies need
to begin preparing for the AV revolution by focusing on not
only the transportation impacts of the technology, but also
the built environment impacts.
Few understood and foresaw the massive impact of the
Model T upon travel behaviors, transportation systems, and
the built environment. Autonomous vehicles will have a simi-
lar significance for our landscapes, and it is our hope that this
remarkable opportunity is grasped and not squandered.

74 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, C H A P T E R 6

www.planning.org AMERICAN PLANNING ASSOCIATION 75

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

REFERENCES

AAA. 2018. “More Americans Willing to Ride in Fully Self-Driving Cars.” Beede, D., R. Powers, and C. Ingram. 2017. The Employment Impact of Au-
AAA Newsroom, January 24. Available at https://newsroom.aaa tonomous Vehicles. Issue Brief #05-17. Washington, D.C.: U.S. Depart-
.com/2018/01/americans-willing-ride-fully-self-driving-cars. ment of Commerce. Available at www.esa.doc.gov/reports/employ-
Abrams, Rachel. 2016. “Self-Driving Cars May Get Here Before We’re ment-impact-autonomous-vehicles.
Ready.” The New York Times, January 21, sec. DealBook. Available at Bertozzi, Massimo, Alberto Broggi, and Alessandra Fascioli. 2000. “Vision-
www.nytimes.com/2016/01/22/business/dealbook/davos-self-driving- Based Intelligent Vehicles: State of the Art and Perspectives.” Robot-
cars-may-get-here-before-were-ready.html. ics and Autonomous Systems 32 (1):1–16. Available at doi.org/10.1016/
Airbib, J., and T. Seba. 2017. Rethinking Transportation 2020-2030: The Dis- S0921-8890(99)00125-6.
ruption of Transportation and the Collapse of the Internal-Combustion Bierstedt, J., A. Gooze, C. Gray, J. Peterman, L. Raykin, and J. Walters. 2014.
Vehicle and Oil Industries. RethinkX. Available at www.rethinkx.com/ “Effects of Next-Generation Vehicles on Travel Demand and Highway
transportation. Capacity.” FP Think. Available at www.fehrandpeers.com/wp-content/
Aitken, N.M., L.G. Pelletier, and D.E. Baxter. 2016. “Doing the Difficult uploads/2015/07/FP_Think_Next_Gen_Vehicle_White_Paper_FI-
Stuff: Influence of Self-Determined Motivation Toward the Environ- NAL.pdf.
ment on Transportation Proenvironmental Behavior.” Ecopsychology Blanco, M., J. Atwood, H.M. Vasquez, T.E. Trimble, V.L. Fitchett, J. Radl-
8(2). Available at doi.org/10.1089/eco.2015.0079. beck, G.M. Fitch, S.M. Russell, C.A. Green, B. Cullinane, and J.F. Mor-
American Public Health Association (APHA). 2009. “Improving Health gan. 2015. Human Factors Evaluation of Level 2 and Level 3 Automated
Through Transportation and Land Use Policies.” November 10. Avail- Driving Concepts. Report No. DOT HS 812 182. Washington, D.C.: Na-
able at www.apha.org/policies-and-advocacy/public-health-policy- tional Highway Traffic Safety Administration. Available at www.nhtsa
statements/policy-database/2014/07/31/08/21/improving-health- .gov/sites/nhtsa.dot.gov/files/812182_humanfactorseval-l2l3-autom-
through-transportation-and-land-use-policies. drivingconcepts.pdf.

Anderson, James M., N. Kalra, K.D. Stanley, P. Sorensen, C. Samaras, and Bomey, N., and C. Woodyard. 2016. “Toyota Signs Ridesharing Deal
O.A. Oluwatola. 2016. Autonomous Vehicle Technology: A Guide for With Uber; VW With Gett.” USA Today, May 24. Available at www
Policymakers. Santa Monica, Calif.: RAND Corporation. Available at .usatoday.com/story/money/cars/2016/05/24/volkswagen-investment-
www.rand.org/pubs/research_reports/RR443-2.html. gett/84851180/

Aon Risk Solutions. 2016. 2016 Public Transit Benchmark Report: Aon Public Bosch, P.M., F. Becker, H. Becker, and K.W. Axhausen. 2017. “Cost-Based
Transit Liability Benchmark Analysis. Executive summary available at Analysis of Autonomous Mobility Services.” Transport Policy 64(May):
www.aon.com/attachments/risk-services/2016-public-transit-liability- 76–91. Available at doi.org/10.1016/j.tranpol.2017.09.005.
analysis-executive-summary.pdf. Boston Consulting Group (BCG). 2015. “Back to the Future: The Road to
Appleyard, Bruce, and William Riggs. 2017. “‘Doing the Right Things’ Rath- Autonomous Driving.” PowerPoint presentation. Available at www.bcg
er than ‘Doing Things Right’: A Conceptual Transportation/Land Use .com/d/press/8jan2015-self-driving-vehicles-market-2025-832.
Framework for Livability, Sustainability and Equity in the Era of Driv- Brant, S. 2016. “Downtown Minneapolis Gets Its First ‘Woonerf.’” Minne-
erless Cars.” SSRN Scholarly Paper ID 3040783. Rochester, N.Y.: Social apolis Star Tribune, September 14. Available at www.startribune.com/
Science Research Network. Available at https://papers.ssrn.com/sol3/ downtown-minneapolis-gets-its-first-woonerf/393491981.
papers.cfm?abstract_id=3040783.

www.planning.org AMERICAN PLANNING ASSOCIATION 77

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

Brown, Austin, Brittany Repac, and Jeff Gonder. 2013. “Autonomous Ve- Crow, A. 2017. “How Safe Are Self-Driving Cars? Waymo Proves They’re
hicles Have A Wide Range of Possible Energy Impacts.” National Re- Pretty Darn Safe.” May 1. Rocky Mountain Institute. Available at http-
newable Energy Laboratory. Workshop on Road Vehicle Automation, www.rmi.org/safe-self-driving-cars/.
Stanford, Calif. Available at www.nrel.gov/docs/fy13osti/59210.pdf. Davidse, R. J. 2006. “Older Drivers and ADAS: Which Systems Improve
Burns, L.D., W. Jordan, and B. Scarborough. 2013. Transforming Personal Road Safety?” IATSS Research 30(1): 6–20. Available at www.science
Mobility, Technical Report. The Earth Institute, Columbia University. direct.com/science/article/pii/S0386111214601515.
Available at http://sustainablemobility.ei.columbia.edu/files/2012/12/ Davies, A. 2017. “General Motors Is Going All Electric.” Wired, October 2.
Transforming-Personal-Mobility-Jan-27-20132.pdf. Available at www.wired.com/story/general-motors-electric-cars-plan-
California, State of, Department of Motor Vehicles. 2017. “Report of Traf- gm.
fic Accident Involving an Autonomous Vehicle.” October 16. Available Daziano, R.A., M.A. Sarrias, and B. Leard. 2017. “Are Consumers Willing to
at www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/autonomous- Pay to Let Cars Drive For Them? Analyzing Response to Autonomous
veh_ol316. Vehicles.” Transportation Research Part C: Emerging Technologies 78:
Cicchino, J.B. 2016. “Effectiveness of Forward Collision Warning Systems 150–64. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_
With and Without Autonomous Emergency Braking in Reducing id=2851943.
Police-Reported Crash Rates.” Arlington, Va.: Insurance Institute for Dennis, E.P., A. Spulber, V.S. Brugeman, R. Kuntzsch, and R. Neuner.
Highway Safety. Available at http://orfe.princeton.edu/~alaink/Smart- 2017. Planning for Connected and Automated Vehicles. Transportation
DrivingCars/Papers/IIHS-CicchinoEffectivenessOfCWS-Jan2016.pdf. Systems Analysis Group. Available at www.cargroup.org/publication/
Chapin, Timothy, Lindsay Stevens, and Jeremy Crute. 2016. “Envisioning planning-for-connected-and-automated-vehicles/.
Florida’s Future: Transportation and Land Use in an Automated Vehicle DiNapoli, T.P., and K.B. Bleiwas. 2008. Metropolitan Transportation Au-
World.” Report prepared for the Florida Department of Transportation. thority: The Cost of Judgements and Claims. Report 2-2009. New York:
Available at https://fpdl.coss.fsu.edu/sites/g/files/imported/storage/ Office of State Comptroller. Available at www.osc.state.ny.us/osdc/
original/application/abfcc477779d0bc0ea825c8011011939.pdf. mta2-2009.pdf.
———. 2017. “Here Come the Robot Cars—Autonomous Vehicles Will Im- Dresner K., and P. Stone. 2008. “A Multiagent Approach to Autonomous
pact the Built Environment: The Time to Plan Is Now.” Planning, April. Intersection Management.” Journal of Artificial Intelligence Research
Available at www.planning.org/planning/2017/apr/robotcars. 21(2008): 591–656. Available at www.aaai.org/Papers/JAIR/Vol31/
Clewlow, Regina R., and Gouri Shankar Mishra. 2017. “Disruptive Trans- JAIR-3117.pdf.
portation: The Adoption, Utilization, and Impacts of Ride-Hailing in the Duncan, M., N. Charness, T. Chapin, M. Horner, L. Stevens, A. Richard, D.
United States.” Research Report UCD-ITS-RR-17-07. Davis, Calif.: UC Souders, J. Crute, A. Riemondy, and D. Morgan. 2015. Enhanced Mobil-
Davis Institute of Transportation Studies. Available at itspubs.ucdavis ity for Aging Populations Using Automated Vehicles. Florida Department
.edu/wp-content/themes/ucdavis/pubs/download_pdf.php?id=2752. of Transportation Report BDV30 977-11. Available at www.fdot.gov/
Cohen, Adam, and Susan Shaheen. 2017. Planning for Shared Mobility. Plan- research/Completed_Proj/Summary_PL/FDOT-BDV30-977-11-rpt.pdf.
ning Advisory Service Report 583. Chicago: American Planning Asso- Edelstein, S. 2018. “Udelv Autonomous Delivery Vehicle Begins Testing
ciation. Available at www.planning.org/publications/report/9107556/. in California.” The Drive, January 31. Available at www.thedrive.com/
Collie, B., J. Rose, R. Choraria, and A.K. Wegscheider. 2017. The Reimagined tech/18113/udelv-autonomous-delivery-vehicle-begins-testing-in-
Car: Shared Autonomous, and Electric. Boston Consulting Group, Decem- california.
ber 18. Available at www.bcg.com/en-us/publications/2017/reimagined- Edwards, J.D., M. Perkins, L.A. Ross, and S.L. Reynolds. 2009. “Driving
car-shared-autonomous-electric.aspx. Status and Three-Year Mortality Among Community-Dwelling Older
Crayton, T.J., and B.M. Meier. 2017. “Autonomous Vehicles: Developing Adults.” The Journals of Gerontology Series A: Biological Sciences and
Public Health Research Agenda to Frame the Future of Transporta- Medical Sciences 64A(2): 300–05. doi:10.1093/gerona/gln019.
tion Policy.” Journal of Transport & Health 6: 245–52. Available at doi Fagnant, Daniel J., and Kara M. Kockelman. 2014. “The Travel and Envi-
.org/10.1016/j.jth.2017.04.004. ronmental Implications of Shared Autonomous Vehicles, Using Agent-
Crothers, Brooke. 2016. “Tesla’s Elon Musk: Car Can Drive Itself Across Based Model Scenarios.” Transportation Research Part C: Emerging
Country in Two Years.” Forbes, January 10. Available at www.forbes Technologies 40:1–13. Available at www.sciencedirect.com/science/
.com/sites/brookecrothers/2016/01/10/teslas-elon-musk-in-two-years- article/pii/S0968090X13002581.
you-can-summon-car-from-across-the-country.

78 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

Fagnant, D.J., K.M. Kockelman, and P. Bansal. 2015. “Operations of a General Motors. 2016. “GM and Lyft to Shape the Future of Mobility.” GM
Shared Autonomous Vehicle Fleet for the Austin, Texas Market.” Trans- Corporate Newsroom, January 4. Available at https://media.gm.com/
portation Research Record 2536: 98–106. Available at http://citeseerx.ist media/us/en/gm/home.detail.html/content/Pages/news/us/en/2016/
.psu.edu/viewdoc/summary?doi=10.1.1.431.1531. Jan/0104-lyft.html.
Fairly, P. 2017. “The Self-Driving Car’s Bicycle Problem.” IEEE Spectrum, Jan- Greenblatt, J.B., and S. Saxena. 2015. “Autonomous Taxis Could Greatly
uary 31. Available at spectrum.ieee.org/cars-that-think/transportation/ Reduce Greenhouse-Gas Emissions of US Light-Duty Vehicles.” Nature
self-driving/the-selfdriving-cars-bicycle-problem. Climate Change 5: 860–62. Available at www.nature.com/articles/ncli-
Federal Highway Administration (FHWA). 2000. Addendum to the 1997 mate2685.
Federal Highway Cost Allocation Study Final Report. Washington, D.C.: Guerra, Erick. 2015a. “When Autonomous Cars Take to the Road.” Plan-
U.S. Department of Transportation. Available at www.fhwa.dot.gov/ ning, May. Available at www.planning.org/planning/2015/may/
policy/hcas/addendum.cfm. autonomouscars.htm.
———. 2014. “Mobility Challenges for Households in Poverty.” FHWA ———. 2015b. “Planning for Cars That Drive Themselves: Metropolitan
NHTS Brief. Available at nhts.ornl.gov/briefs/PovertyBrief.pdf. Planning Organizations, Regional Transportation Plans, and Autono-
Federal Transit Administration (FTA). 2017. Strategic Transit Automa- mous Vehicles.” Journal of Planning Education and Research 36(2):
tion Research Plan Roadmap. Washington, D.C.: U.S. Department of 210–24. Available at doi.org/10.1177/0739456X15613591.
Transportation. Available at www.transit.dot.gov/sites/fta.dot.gov/files/ Hamari, J., M. Sjöklint, and A. Ukkonen. 2015. “The Sharing Economy: Why
strategic-transit-automation-research-roadmap.PDF. People Participate In Collaborative Consumption.” Journal of the Asso-
Ford Motor Company. 2015. “Ford at CES Announces Smart Mobility Plan ciation for Information Science and Technology 67(9): 2047–59. Available
and 25 Global Experiments Designed to Change the Way the World at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2271971.
Moves.” Ford Media Center, January 6. Available at https://media.ford Harb, M., Y. Xiao, G. Circella, P. Mokhtarian, and J. Walker. 2018. “Project-
.com/content/fordmedia/fna/us/en/news/2015/01/06/ford-at-ces- ing Travelers into a World of Self-Driving Cars: Naturalistic Experiment
announces-smart-mobility-plan.html. for Travel Behavior Implications.” In Proceedings of the 97th Transporta-
Fielding, Gordon J., Timlynn T. Babitsky, and Mary E. Brenner. 1985. “Per- tion Research Board Annual Meeting. Washington, D.C.: Transportation
formance Evaluation for Bus Transit.” Transportation Research Part Research Board. Available at https://trid.trb.org/view/1497254.
A: General 19(1): 73–82. Available at www.sciencedirect.com/science/ Harris, Mark. 2015. “Documents Confirm Apple Is Building Self-Driving
article/pii/0191260785900093. Car.” The Guardian, August 14. Available at www.theguardian.com/
Fielding, Gordon J., Roy E. Glauthier, and Charles A. Lave. 1978. “Perfor- technology/2015/aug/14/apple-self-driving-car-project-titan-sooner-
mance Indicators for Transit Management.” Transportation 7 (4): 365– than-expected.
79. Available at doi.org/10.1007/BF00168037. Henaghan, Jennifer, David C. Rouse, Kelley Coyer, Lisa Nisenson, and
Friedrich, B. 2016. “The Effect of Autonomous Vehicles on Traffic.” In M. Jason Jordan. 2018. Preparing Communities for Autonomous Vehicles.
Maurer, C. Gerdes, B. Lenz, and H. Winner, Autonomous Driving, pp. Chicago: American Planning Association. Available at www.planning
317–34. Available at doi 10.1007/978-3-662-48847-8_16. .org/publications/document/9144551.

Fulton, Lew, Jacob Mason, and Dominique Meroux. 2017. Three Revolutions Hidalgo, J. 2017. “Flirtey, 7-Eleven Complete 77 On-Demand Drone Deliveries
in Urban Transportation. UC Davis Institute of Transportation Stud- in Reno.” Technobubble, Reno Gazette Journal, July 26. Available at www
ies and Institute for Transportation & Development Policy. Available at .rgj.com/story/money/business/2016/12/21/flirtey-7-eleven-completes-
www.itdp.org/publication/3rs-in-urban-transport/. 77-drone-deliveries-reno-technobubble/95714592.

Gardner, Greg. 2016. “Why Most Self-Driving Cars Will Be Electric.” USA Huang, J., and H. Tan. 2016. “Control System Design of an Automated Bus
Today, September 19. Available at www.usatoday.com/story/money/ in Revenue Service.” IEEE Transactions on Intelligent Transportation
cars/2016/09/19/why-most-self-driving-cars-electric/90614734. Systems, 10: 2868. Available at https://ieeexplore.ieee.org/document/
7464847.
Gayah, Vikash V., and Carlos F. Daganzo. 2012. “Analytical Capacity Com-
parison of One-Way and Two-Way Signalized Street Networks.” Trans- International Transport Forum. (ITF) 2017. Managing the Transition to
portation Research Record: Journal of the Transportation Research Board Driverless Road Freight Transport: Case Specific Policy Analysis. Avail-
2301(1): 76–85. Available at https://doi.org/10.3141/2301-09. able at www.itf-oecd.org/sites/default/files/docs/managing-transition-
driverless-road-freight-transport.pdf.

www.planning.org AMERICAN PLANNING ASSOCIATION 79

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

Jariyasunant, Jerald, Maya Abou-Zeid, Andre Carrel, Venkatesan Ekam- Lanctot, Roger. 2017. Accelerating the Future: The Economic Impact of
baram, David Gaker, Raja Sengupta, and Joan L. Walker. 2015. “Quan- the Emerging Passenger Economy. Strategy Analytics. June. Avail-
tified Traveler: Travel Feedback Meets the Cloud to Change Behavior.” able at https://newsroom.intel.com/newsroom/wp-content/uploads/
Journal of Intelligent Transportation Systems 19(2): 109–24. Available at sites/11/2017/05/passenger-economy.pdf.
doi.org/10.1080/15472450.2013.856714. Larco, Nico. 2017. “Micro-Transit Is Not a Free-for-All.” Urbanism Next,
Johnson, Brian. 2015. “Disruptive Mobility.” July 20. Barclays. Available July 10. Available at https://urbanismnext.uoregon.edu/2017/07/10/
at https://orfe.princeton.edu/~alaink/SmartDrivingCars/PDFs/Brian_ micro-transit-is-not-a-free-for-all/.
Johnson_DisruptiveMobility.072015.pdf. Lioris, Jennie, Ramtin Pedarsani, Fatma Tascikaraoglu, and Pravin Vara-
Johnson, Charlie, and Jonathan Walker. 2017. Peak Car Ownership: The iya. 2017. “Platoons of Connected Vehicles Can Double Throughput in
Market Opportunity of Electric Automated Mobility Services. Boulder, Urban Roads.” Journal of Transportation Research Part C: Emerging
Colo.: Rocky Mountain Institute. Available at https://rmi.org/insight/ Technologies 77(April): 292–305. Available at https://doi.org/10.1016/
peak-car-ownership-report. j.trc.2017.01.023.
Juniper Research. 2017. “Sharing Economy: Market Snapshot 2017.” Avail- Litman, Todd. 2018. Autonomous Vehicle Implementation Predictions: Im-
able at www.juniperresearch.com/resources/infographics/sharing- plications for Transport Planning. Victoria, B.C.: Victoria Transport
economy-market-snapshot-2017. Policy Institute. Available at www.vtpi.org/avip.pdf.
Kelly, Alonzo, Anthony Stentz, Omead Amidi, Mike Bode, David Bradley, Lunden, Ingrid. 2016. “Uber Says That 20% of Its Rides Globally Are
Antonio Diaz-Calderon, Mike Happold, et al. 2006. “Toward Reliable Now on UberPool.” TechCrunch, May 10. Available at techcrunch
Off Road Autonomous Vehicles Operating in Challenging Environ- .com/2016/05/10/uber-says-that-20-of-its-rides-globally-are-now-on-
ments.” The International Journal of Robotics Research 25(5–6): 449–83. uber-pool.
Available at doi.org/10.1177/0278364906065543. Marakby, Sherif. 2018. “We’re Going To Miami: The First Proving Ground
Kockelman, K., S. Boyles, P. Stone, D. Fagnant, R. Patel, M.W. Levin, G. Sha- For Our Self-Driving Service.” Press Release, Ford Media Center, Feb-
ron, et al. 2017. An Assessment of Autonomous Vehicles: Traffic Impacts ruary 27. Available at https://media.ford.com/content/fordmedia/fna/
and Infrastructure Needs. Report No. FHWA/TX-17/0-6847-1. Austin: us/en/news/2018/02/27/we_re-going-to-miami--the-first-proving-
The University of Texas at Austin Center for Transportation Research. ground-for-our-self-driv.html.
Available at www.cs.utexas.edu/~guni/Papers/TxDOT-17.pdf. Marottoli, R.A., C.F. Mendes, T.A. Glass, C.S. Williams, L.M. Cooney, L.F.
Kodransky, M. 2016. “Driverless Cars Will Not Solve Inequal- Berkman, and M.E. Tinnetti. 1997. “Driving Cessation and Increased
ity in Cities.” Sustainable Transport, Winter. Available at Depressive Symptoms: Prospective Evidence from the New Haven
https://3gozaa3x xbpb499ejp30lxc8-wpengine.netdna-ssl.com/ EPESE.” Journal of the American Geriatrics Society 45(2): 202–6. Avail-
wp-content/uploads/2016/01/Driverless-Cars-a-debate.pdf. able at www.ncbi.nlm.nih.gov/pubmed/9033520.
Kostyniuk, L.P., D.A. Trombley, and J.T. Shope. 1998. The Process of Re- Marshall, Aarian. 2017. “To Let Self-Driving Cars Go Anywhere, You Must
duction and Cessation of Driving among Older Drivers: A Review of Train Them Everywhere.” Wired, October 27. Available at www.wired
the Literature. The University of Michigan Transportation Research .com/story/waymo-self-driving-michigan-testing.
Institute. Available at https://deepblue.lib.umich.edu/bitstream/ McDowell, Mark. 2014. “Time for Autonomous Vehicles to Disrupt Trans-
handle/2027.42/1244/91238.0001.001.pdf?sequence=2&isAllowed=y. portation Planning.” Eno Transportation Weekly. Available at www
Kuhr, James, Natalia Ruiz Juri, Chandra Bhat, Jackson Archer, Jennifer Duth- .enotrans.org/eno-brief/time-for-autonomous-vehicles-to-disrupt-
ie, Edgar Varela, Maitri Zalawadia, Thomas Bamonte, Arash Mirzaei, transportation-planning#_ednref13.
and Hong Zheng. 2017. Travel Modeling in an Era of Connected and Auto- McFarland, Matt. 2015. “What Happens When Cities and States Try to Pre-
mated Transportation Systems: An Investigation in the Dallas-Fort Worth pare for Self-Driving Cars.” Washington Post, November 25. Available
Area. Report No. D-STOP/2017/122. North Central Texas Council of at www.washingtonpost.com/news/innovations/wp/2015/11/25/what-
Governments. Available at ctr.utexas.edu/wp-content/uploads/122.pdf. happens-when-cities-and-states-try-to-prepare-for-self-driving-cars/.
Kung, K.S., K. Greco, S. Sobolevsky, and C. Ratti. 2014. “Exploring Univer- McKinsey & Company. 2016. Automotive Revolution—Perspective Towards
sal Patterns in Human Home-Work Commuting from Mobile Phone 2030: How the Convergence of Disruptive Technology-Driven Trends
Data.” PLOS ONE 9(6): e96180. Available at http://journals.plos.org/ Could Transform the Auto Industry. January. Available at www.mckinsey
plosone/article?id=10.1371/journal.pone.0096180. .com/industries/automotive-a nd-assembly/our-insights/dis-
ruptive-trends-that-will-transform-the-auto-industr y/de-de.

80 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

———. 2017. Electrifying Insights: How Automakers Can Drive Electrified Nagata, Kazuaki. 2018. “Yamato and DeNA Test Autonomous Delivery Sys-
Vehicle Sales And Profitability. January. Available at www.mckinsey tem in Japan.” The Japan Times, April 24. Available at www.japantimes
.com/industries/automotive-and-assembly/our-insights/electrifying- .co.jp/news/2018/04/24/business/tech/yamato-dena-test-autonomous-
insights-how-automakers-can-drive-electrified-vehicle-sales-and- delivery-system-japan/#.Wuf-34jwbIX.
profitability. National Association of City Transportation Officials (NACTO). 2016.
Messner, Matthew. 2017. “A Michigan Facility Is the Nation’s Epicenter for “NACTO Policy Statement on Automated Vehicles.” Available at nacto
Testing Self-Driving Cars.” Architect’s Newspaper, October 23. Avail- .org/wp-content/uploads/2016/06/NACTO-Policy-Automated-
able at https://archpaper.com/2017/10/autonomous-vehicles-michigan- Vehicles-201606.pdf.
willow-run. ———. 2017. “Curb Appeal: Curbside Management Strategies for Improv-
Milakis, Dimitris, Bart van Arem, and Bert van Wee. 2017. “Policy and So- ing Transit Reliability.” NACTO Transit Leadership Resource Paper,
ciety Related Implications of Automated Driving: A Review of Litera- November. Available at https://nacto.org/wp-content/uploads/2017/11/
ture and Directions for Future Research.” Journal of Intelligent Trans- NACTO-Curb-Appeal-Curbside-Management.pdf.
portation Systems 21(4): 324–48. Available at https://doi.org/10.1080/15 National Highway Traffic Safety Administration (NHTSA). 2015. The Eco-
472450.2017.1291351. nomic and Societal Impact of Motor Vehicle Crashes, 2010 (Revised).
Milam, R. and Riggs, W. Forthcoming. “Balancing Promise with Peril.” DOT Publication No. HS 812 013. Washington, D.C.: National Highway
In Disruptive Transport: Driverless Cars, Transport Innovation and Traffic Safety Administration. Available at https://crashstats.nhtsa.dot
the Sustainable City of Tomorrow, edited by William Riggs. London: .gov/Api/Public/ViewPublication/812013.
Routledge. Available at www.routledge.com/Disruptive-Transport- ———. 2016. Federal Automated Vehicles Policy: Accelerating the Next Revo-
Driverless-Cars-Transport-Innovation-and-the-Sustainable/Riggs/p/ lution in Roadway Safety. U.S. Department of Transportation. Available
book/9781138613164. at www.transportation.gov/sites/dot.gov/files/docs/AV%20policy%20
Minneapolis, City of, Department of Public Works. 2010. “Hennepin Av- guidance%20PDF.pdf.
enue and 1st Avenue Two-Way Conversion Evaluation Report.” Avail- ———. 2017a. “2016 Fatal Motor Vehicle Crashes: Overview.” Traffic Safety
able at www.minneapolismn.gov/www/groups/public/@publicworks/ Facts Research Note, October. DOT HS 812 456. Washington, D.C.:
documents/webcontent/convert_276126.pdf. National Highway Traffic Safety Administration. Available at https://
Morgan Stanley. 2013. “Autonomous Cars: Self-Driving the New Auto crashstats.nhtsa.dot.gov/Api/Public/Publication/812456.
Industry Paradigm.” Morgan Stanley Blue Paper, November 6. Avail- ———. 2017b. Automated Driving Systems 2.0: A Vision for Safety. Report
able at https://orfe.princeton.edu/~alaink/SmartDrivingCars/PDFs/ No. DOT HS 812 442.Washington, D.C.: National Highway Traffic Safe-
Nov2013MORGAN-STANLEY-BLUE-PAPER-AUTONOMOUS- ty Administration. Available at https://www.nhtsa.gov/sites/nhtsa.dot
CARS%EF%BC%9A-SELF-DRIVING-THE-NEW-AUTO-INDUS- .gov/files/documents/13069a-ads2.0_090617_v9a_tag.pdf.
TRY-PARADIGM.pdf.
National Transportation Safety Board (NTSB). 2017. Collision Between a
——— 2016. “Shared Mobility on the Road of the Future.” Morgan Stanley Car Operating with Automated Vehicle Control Systems and a Tractor-
Research, June 15. Available at www.morganstanley.com/ideas/car-of- Semitrailer Truck Near Williston, Florida May 7, 2016. NTSB/HAR-
future-is-autonomous-electric-shared-mobility. 17-02. Washington, D.C. Available at www.ntsb.gov/investigations/
Mosquet, X, T. Dauner, N. Land, M. Rubmann, A. MeiPochtler, R. Agraw- AccidentReports/Reports/HAR1702.pdf.
al, and F. Schmieg. 2015. “Revolution in the Driver’s Seat: The Road to New York, City of. 2017. “Mayor de Blasio Announces Initiatives to Help
Autonomous Vehicles.” April 21. Boston Consulting Group. Available Ease Congestion.” Office of the Mayor, October 22. Available at https://
at www.bcg.com/publications/2015/automotive-consumer-insight- www1.nyc.gov/office-of-the-mayor/news/673-17/mayor-de-blasio-
revolution-drivers-seat-road-autonomous-vehicles.aspx#chapter1. initiatives-help-ease-congestion#/0.
Mousavian, Arsalan, Dragomir Anguelow, John Flynn, and Jana Kosec- Quarles, Neil, and Kara Kockelman. 2018. “Costs and Benefits of Electri-
ka. 2017. “3D Bounding Box Estimation Using Deep Learning and fying and Automating U.S. Bus Fleets.” Paper Presented at the 97th
Geometry.” IEEE Spectrum. Available at http://openaccess.thecvf Annual Meeting of the Transportation Research Board. Washington,
.com/content_cvpr_2017/papers/Mousavian_3D_Bounding_Box_ D.C. Available at www.ce.utexas.edu/prof/kockelman/public_html/
CVPR_2017_paper.pdf. TRB18AeBus.pdf.

www.planning.org AMERICAN PLANNING ASSOCIATION 81

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

Richland, Judi, Jim Lee, and Erin Duggan Butto. 2016. “Steering Au- Schlossberg, Marc, William Riggs, Adam Millard-Ball, and Eliza-
tonomous Vehicle Policy: The Role of Public Health.” Health Af- beth Shay. 2018. Rethinking the Street in an Era of Driverless Cars.
fairs Blog, March 4. Available at www.healthaffairs.org/do/10.1377/ UrbanismNext. Available at https://urbanismnext.uoregon.edu/
hblog20160304.053688/full. files/2018/01/Rethinking_Streets_AVs_012618-27hcyr6.pdf.
Riggs, William. Forthcoming. Disruptive Transport: Driverless Cars, Schoettle, Brandon, and Michael Sivak. 2014. A Survey of Public Opinion
Transport Innovation and the Sustainable City of Tomorrow. London: About Autonomous and Self-Driving Vehicles in the U.S., the UK., and
Routledge. Available at www.routledge.com/Disruptive-Transport- Australia. UMTRI-2014-21. Ann Arbor: The University of Michigan
Driverless-Cars-Transport-Innovation-and-the-Sustainable/Riggs/p/ Transportation Research Institute. Available at https://deepblue.lib
book/9781138613164. .umich.edu/handle/2027.42/108384.
———. 2017. “Painting the Fence: Social Norms as Economic Incentives ———. 2015. Potential Impact of Self-Driving Vehicles on Household Vehicle
to Non-Automotive Travel Behavior.” Travel Behaviour and Society Demand and Usage. UMTRI-2015-3. Ann Arbor: The University of Mich-
7(April): 26–33. Available at doi.org/10.1016/j.tbs.2016.11.004. igan Transportation Research Institute. Available at https://deepblue
Riggs, William, and Michael R. Boswell. 2016a. “Thinking Beyond the (Au- .lib.umich.edu/handle/2027.42/110789.
tonomous) Vehicle: The Promise of Saved Lives.” Planetizen, December Scott, Mark. 2017. “The Future of European Transit: Driverless and Utili-
13. Available at https://works.bepress.com/williamriggs/71/. tarian.” The New York Times, May 28. Available at www.nytimes
———. 2016b. “No Business as Usual in an Autonomous Vehicle Fu- .com/2017/05/28/technolog y/the-future-of-european-transit-
ture.” Planetizen, March 28. Available at https://works.bepress.com/ driverless-and-utilitarian.html.
williamriggs/53. Smolnicki, Piotr Marek, and Jacek Sołtys. 2016. “Driverless Mobility: The
———. 2016c. “The Bicyclists’ Manifesto for an Autonomous Vehicle Fu- Impact on Metropolitan Spatial Structures.” Procedia Engineering,
ture.” Planetizen, September 23. Available at https://works.bepress World Multidisciplinary Civil Engineering-Architecture-Urban Plan-
.com/williamriggs/62/. ning Symposium 2016, WMCAUS 2016, 161 (Supplement C):2184–90.
Available at doi.org/10.1016/j.proeng.2016.08.813.
Riggs, William, and John Gilderbloom. 2016a. “Two-Way Street Con-
version: Evidence of Increased Livability in Louisville.” Journal of Society of Automotive Engineers (SAE) International. 2016. “U.S. Depart-
Planning Education and Research 36(1): 105–18. Available at doi ment of Transportation’s New Policy on Automated Vehicles Adopts
.org/10.1177/0739456X15593147. SAE International’s Levels of Automation for Defining Driving Auto-
mation in On-Road Motor Vehicles.” Press release.
———. 2016b. “The Connection between Neighborhood Walkability and Life
Longevity in a Midsized City.” Focus 13(1). Available at digitalcommons Sorrel, Charlie. 2017. “Bikes Are Still a Headache for Self-Driving Cars.” Fast
.calpoly.edu/focus/vol13/iss1/11. Company, February 27. Available at www.fastcompany.com/3068141/
bikes-are-still-a-headache-for-self-driving-cars.
———. 2017. “How Multi-Lane, One-Way Street Design Shapes Neigh-
bourhood Life: Collisions, Crime and Community.” Local Envi- Stern, Raphael E., Shumo Cui, Maria Laura Delle Monache, Rahul Bhadani,
ronment 22(8): 917–33. Available at www.tandfonline.com/eprint/ Matt Bunting, Miles Churchill, Nathaniel Hamilton, R’mani Haulcy,
qZNFa3D9zyn9RD5chhvE/full. Hannah Pohlmann, Fangyu Wu, Benedetto Piccoli, Benjamin Seibold,
Jonathan Sprinkle, and Daniel B. Work. 2018. “Dissipation of Stop-and-
Rosen, Ben. 2016. “Google Is Teaching Its Self-Driving Cars to Share the
Go Waves Via Control of Autonomous Vehicles: Field Experiments.”
Road with Cyclists.” The Christian Science Monitor, July 6. Available at
Transportation Research Part C: Emerging Technologies 89: 205–21.
www.csmonitor.com/Technology/2016/0706/Google-is-teaching-its-
Available at arxiv.org/abs/1705.01693.
self-driving-cars-to-share-the-road-with-cyclists.
Talebpour, Alireza, and Hani S. Mahmassani. 2016. ”Influence of Connect-
Rosenbloom, Sandra. 2003. “The Mobility Needs of Older Americans: Im-
ed And Autonomous Vehicles on Traffic Flow Stability and Through-
plications for Transportation Reauthorization.” Transportation Reform
put.” Transportation Research Part C: Emerging Technologies 71: 143–63.
Series, July. Washington, D.C.: Brookings Institution. Available at www
Available at doi.org/10.1016/j.trc.2016.07.007.
.brookings.edu/research/the-mobility-needs-of-older-americans-
implications-for-transportation-reauthorization/. Tampa Hillsborough Expressway Authority. 2017. “Tampa Connected Ve-
hicle Pilot.” Available at www.tampacvpilot.com.
San Luis Obispo (California), City of. 2017. “Circulation.” Chapter 2 in City
of San Luis Obispo General Plan. Available at www.slocity.org/home/ Tesla. 2015. “Your Autopilot Has Arrived.” Tesla Blog, October 14. Available
showdocument?id=20412. at www.tesla.com/blog/your-autopilot-has-arrived.

82 AMERICAN PLANNING ASSOCIATION www.planning.org

 PLANNING FOR AUTONOMOUS MOBILITY
PA S 592, R E F E R E N C E S

Thompson,Cadie.2017.“MercedesCreatedtheBusoftheFuture—AndItLooks Yaraghi, Niam, and Shamika Ravi. 2017. The Current and Future State of
Awesome.” Business Insider, June 11. Available at www.businessinsider the Sharing Economy. Impact Series NO. 032017, March. Chanakyapuri,
.com/mercedes-future-bus-city-pilot-photos-features-2017-6. New Delhi: The Brookings Institution India. Available at www.brookings
Tonguz, Ozan K. 2011. “Biologically Inspired Solutions to Fundamental .edu/wp-content/uploads/2016/12/sharingeconomy_032017final.pdf.
Transportation Problems.” IEEE Communications Magazine 49(11): Yassuda, Monica S., Jennifer J. Wilson, and Otto von Mering. 1997.
106–15. Available at https://ieeexplore.ieee.org/document/6069717. “Driving Cessation: The Perspective of Senior Drivers.” Educational
Underwood, Steve. 2015. Automated, Connected, and Electric Vehicle Sys- Gerontology 23(6): 525–38. Available at www.tandfonline.com/doi/
tems: Expert Forecast and Roadmap for Sustainable Transportation. abs/10.1080/0360127970230603.
Dearborn: University of Michigan Institute for Advanced Vehicle Sys- Zimmer, John, and Logan Green. 2018. “Lyft’s Approach To Bikes & Scoot-
tems. Available at http://graham.umich.edu/media/files/LC-IA-ACE- ers: Accelerating Public + Private Mobility.” Medium, July 16. Avail-
Roadmap-Expert-Forecast-Underwood.pdf. able at https://medium.com/@johnzimmer/lyfts-approach-to-bikes-
U.S. Department of Transportation (USDOT). 2016. “Connected Vehicle scooters-90ce505ff496.
Pilot Deployment Program.” Available at www.its.dot.gov/pilots.
———.2018. “U.S. Department of Transportation Announces Release of Au-
tomated Vehicle Requests for Public Comment.” January 10. Available
at www.transportation.gov/briefing-room/dot0518.
U.S. Environmental Protection Agency (EPA). 2017. “Sources of Greenhouse
Gas Emissions: Transportation Sector Emissions.” Available at www.epa
.gov/ghgemissions/sources-greenhouse-gas-emissions#transportation.
Wadud, Zia, Don MacKenzie, and Paul Leiby. 2016. “Help or Hindrance?
The Travel, Energy and Carbon Impacts of Highly Automated Vehicles.”
Transportation Research Part A: Policy and Practice 86: 1–18. Available
at doi.org/10.1016/j.tra.2015.12.001.
Walker Consultants. 2017. “Timing of AVs And Transportation Disrup-
tion.” Available at www.walkerconsultants.com/wp-content/uploads/
2017/12/Timing-of-AVs.pdf.
Waymo. 2017a. Waymo Safety Report: On the Road To Fully Self-Driving.
Available at waymo.com/safetyreport.
———. 2017b. Disengagement Report: Report on Autonomous Mode Dis-
engagements for Waymo Self-Driving Vehicles in California. Avail-
able at www.dmv.ca.gov/portal/wcm/connect/946b3502-c959-4e3b-
b119-91319c27788f/GoogleAutoWaymo_disengage_report_2016
.pdf?MOD=AJPERES.
White House Office of the Press Secretary. 2015. “Fact Sheet: Administra-
tion Announces New “Smart Cities” Initiative to Help Communities
Tackle Local Challenges and Improve City Services.” Press Release,
September 14. Available at https://obamawhitehouse.archives.gov/the-
press-office/2015/09/14/fact-sheet-administration-announces-new-
smart-cities-initiative-help.
World Health Organization (WHO). 2015. Global Statius Report on Road
Safety 2015 Summary. France: World Health Organization. Available
at www.who.int/violence_injury_prevention/road_safety_status/2015/
en/.

www.planning.org AMERICAN PLANNING ASSOCIATION 83

 DESIGN REVIEW: GUIDING BETTER DEVELOPMENT
PA S 591, AC K N O W L E D G M E N T S

ACKNOWLEDGMENTS

The authors would like to acknowledge and thank all those who contributed to the writing and creation of this report. In
particular, we would like to thank David Dixon, faia; Erick Guerra, phd; and Lisa Nisenson for reviewing the early drafts and
providing insightful feedback to strengthen the report. David Dixon and Stantec’s Urban Places deserve special thanks for
providing the cover image for this report. We would also like to thank Kathy Makinen of the DeVoe Moore Center at Florida
State University for expert copyediting assistance.
Chapter 4 of this report draws heavily from a Florida Department of Transportation-funded studio project led by then-
Florida State University Planner-in-Residence Lindsay Stevens, aicp, entitled Envisioning Florida’s Future: Transportation and
Land Use in an Automated Vehicle World. She was assisted on the project by Tim Chapin and Jeremy Crute, also of Florida
State University. The authors are particularly appreciative of Ed Hutchinson at the Florida Department of Transportation and
Tanner Martin, a private consultant, who played key roles in supporting that project and advancing the role of planning in the
rollout of autonomous vehicle technology in Florida. Special thanks are also due to Jordan Crandall, Mark Horner, Michael
Duncan, Alexandria Washington, Anne Rokyta, aicp, and the Florida State University Department of Urban & Regional
Planning Fall 2015 Studio Team for their contributions to the Envisioning Florida’s Future project.
Similarly, many of the ideas in Chapter 5 have some amount of origination in the Autonomous Future series on Planeti-
zen.com by William Riggs, aicp, and Michael Boswell, aicp; the University of San Francisco’s 2017 Autonomous Vehicles &
the City Symposium (sponsored by the University of San Francisco, Cal Poly San Luis Obispo, Mineta Transportation Institute
and Arup); Transportation Research Board work from Bruce Appleyard and William Riggs; and from research partnerships
with the Urbanism Next team at the University of Oregon. We are grateful for the ideation that stemmed from each of these
teams as we developed this document.
Finally, a very special thanks to Ann Dillemuth, aicp; David Rouse, faicp; and APA’s PAS team for their invaluable as-
sistance, expertise, and patience throughout the writing and development process. We greatly appreciate their help keeping us
on track, editing and formatting the report, and making the process as smooth as possible.

84 AMERICAN PLANNING ASSOCIATION www.planning.org

MEMBERSHIP
Now includes PAS publications!
All APA members now get digital access to every new Planning
Advisory Service publication—each one filled with expert guidance
on big planning challenges, relevant research, and best practices.

Previously available only to subscribers, these authoritative resources

are now included with APA membership.

Digital PAS publications include:

PAS Reports I PAS Memo I PAS QuickNotes
PAS Essential Info Packets

And, there’s more! Members also get unlimited access to the

entire PAS online archive. Hundreds of resources are available for
download—right now.

APA membership. Always good. Now even better.

Learn more at planning.org/pas

American Planning Association
205 N. Michigan Ave., Suite 1200
Chicago, IL 60601-5927

planning.org