Software Engineering for Automotive Systems: Principles and Applications 1st Edition P. Sivakumar (Editor) pdf download
Software Engineering for Automotive Systems: Principles and Applications 1st Edition P. Sivakumar (Editor) pdf download
https://ebookmeta.com/product/software-engineering-for-
automotive-systems-principles-and-applications-1st-edition-p-
sivakumar-editor/
https://ebookmeta.com/product/systems-engineering-for-automotive-
powertrain-development-hannes-hick-editor/
https://ebookmeta.com/product/automotive-embedded-systems-key-
technologies-innovations-and-applications-m-kathiresh/
https://ebookmeta.com/product/sustainable-development-goals-and-
migration-1st-edition-p-sivakumar-editor/
https://ebookmeta.com/product/harmony-from-discords-a-life-of-
sir-john-denham-brendan-o-hehir/
Memory and Religion from a Postsecular Perspective 1st
Edition Zuzanna Bogumi■ Editor Yuliya Yurchuk Editor
https://ebookmeta.com/product/memory-and-religion-from-a-
postsecular-perspective-1st-edition-zuzanna-bogumil-editor-
yuliya-yurchuk-editor/
https://ebookmeta.com/product/infrastructure-development-and-
construction-management-1st-edition-j-c-edison/
https://ebookmeta.com/product/business-for-communicators-the-
essential-guide-to-success-in-corporate-and-public-affairs-1st-
edition-duhe/
https://ebookmeta.com/product/the-haunting-of-pemberton-
manor-1st-edition-damien-dark/
https://ebookmeta.com/product/learn-python-by-coding-video-games-
intermediate-a-step-by-step-guide-to-coding-in-python-fast-
patrick-felicia/
Essential Echocardiography: A Review of Basic
Perioperative TEE and Critical Care Echocardiography
Timothy M. Maus (Editor)
https://ebookmeta.com/product/essential-echocardiography-a-
review-of-basic-perioperative-tee-and-critical-care-
echocardiography-timothy-m-maus-editor/
Software Engineering for
Automotive Systems
Software Engineering for
Automotive Systems
Edited by
P. Sivakumar, B. Vinoth Kumar, and R. S. Sandhya Devi
First edition published 2022
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742
Reasonable efforts have been made to publish reliable data and information, but the author and pub-
lisher cannot assume responsibility for the validity of all materials or the consequences of their use.
The authors and publishers have attempted to trace the copyright holders of all material reproduced
in this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so
we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, trans-
mitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.com
or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-
750-8400. For works that are not available on CCC please contact [email protected]
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are
used only for identification and explanation without intent to infringe.
DOI: 10.1201/9781003269908
Typeset in Times
by KnowledgeWorks Global Ltd.
Contents
Preface......................................................................................................................vii
Editor Biographies.....................................................................................................ix
v
vi Contents
vii
viii Preface
We are grateful to the authors and reviewers for their excellent contributions in
making this book possible.
We are grateful to CRC Press and Taylor and Francis Group, especially to
Gauravjeet Singh Reen (Senior Commissioning Editor) for the excellent collabora-
tion and support.
This edited book covers the theory and case studies of software engineering for
automotive systems. We hope the chapters presented will inspire researchers and
practitioners from academia and industry to spur further advances in the field.
Dr P. Sivakumar
Dr B. Vinoth Kumar
Ms R. S. Sandhya Devi
July 2021
Editor Biographies
P. Sivakumar received his B.E. degree in Electrical and
Electronics with I class in 2006 from Anna University. He
completed his M.E. degree in Embedded System
Technologies with I class in 2009 from Anna University
Coimbatore. He completed his Ph.D. in Electrical Engineering
with a specialization in Automotive Embedded Software in
2018 from Anna University, PSG College of Technology. His
research interests include embedded systems, model-based
design, model-based testing of automotive software, automo-
tive software development, fog and edge computing. He serves as a Guest Editor/
Reviewer of journals with leading publishers such as Inderscience and Springer.
ix
1 Role of AUTOSAR
in Automotive
Software Trends
P. Sivakumar and A. Pavithra
Department of EEE, PSG College of Technology,
Coimbatore, India
S. K. Somasundarum
Department of IT, PSG College of Technology,
Coimbatore, India
P. K. Somanathan
ZF Friedrichshafen AG, Solihull, England,
United Kingdom
A. Manimuthu
Cybersecurity Centre @ NTU (CYSREN), Nanyang
Technological University, Singapore
CONTENTS
1.1 Introduction....................................................................................................... 2
1.1.1 Key Drivers............................................................................................ 3
1.1.2 Key Restraint......................................................................................... 4
1.2 Classic Platform................................................................................................. 4
1.3 Adaptive Platform..............................................................................................4
1.4 Market for Automotive Software and Core Technologies.................................5
1.4.1 Consolidation of ECUs.......................................................................... 5
1.4.2 Updates Received OTA......................................................................... 5
1.4.3 Role of Artificial Intelligence (AI) in Automotive................................ 5
1.4.4 Mission of AUTOSAR..........................................................................5
1.4.5 Purpose of AUTOSAR.......................................................................... 6
1.4.6 Methodology..........................................................................................6
1.5 AUTOSAR’s Highlights.................................................................................... 6
1.5.1 Enhance Security and Safety................................................................. 6
1.5.2 Connectivity Should Be Improved........................................................6
1.5.3 Develop Service Life Enhancements That Are Adaptable....................6
DOI: 10.1201/9781003269908-1 1
2 Software Engineering for Automotive Systems
1.1 INTRODUCTION
Automotive open system architecture (AUTOSAR) is a plug-and-play platform
architecture that allows vehicle original equipment manufacturers (OEMs) and
Tier 1 provider to increase electronic control unit (ECU) programming efficiency,
reduce development costs, and prevent re-advancement of identical ECU software
program elements for the same vehicle applications. It is a new and evolving way
of describing a layered programming system. AUTOSAR aims to enhance com-
plexity management of built-in E/E architectures by allowing OEMs and vendors
to reuse and swap software modules (Devi, R. S., Sivakumar, P., & Balaji, R.
2018). AUTOSAR aims to standardize the structure of ECU. AUTOSAR sets the
stage for cutting-edge digital systems to improve efficiency, safety, and protection,
among other things.
Automotive attributes such as independent driving, vehicle-to-everything
(V2X) availability, over-the-air (OTA) refreshes, prescient support, and a wide
range of current focuses are fundamentally founded on in-vehicle programming
capacities. Each ECU must be functional in order for each of these features to
operate flawlessly and to take into account continuous in-vehicle capabilities. To
help complicated vehicular tasks, cutting-edge top-of-the-line engines have up
to a hundred ECUs that communicate with each other using Ethernet, Controller
Area Networking (CAN), FlexRay, CAN Flexible Data-Rate (FD), and other
Role of AUTOSAR in Automotive Software Trends 3
In 2013, the AUTOSAR group launched an uninterrupted working mode for its
traditional model in order to maintain the well-known while providing selected
upgrades. The fundamental advancement exercises had been posted along these lines
in some kind of a shared arrival of AUTOSAR classic, foundation, and adaptive,
with release 18.10 in October 2018.The automotive software program is a collection
of instructible data development that is used to carry out computerized in-vehicle
operations. In-vehicle microcontroller’s software is often referred to as automo-
tive software (Sivakumar, P., Vinod, B., Devi, R. S. 2016). Telemetry, multimedia,
power train, body manipulation and convenience, connectivity, and enhanced driver
assistance systems (EDAS) are examples of computerized in-vehicle implementa-
tions. Automotive software refers to the hardware and interfaces that run in-vehicle-
integrated functions walkthrough on a machine (Mahmud, N. et al 2018) (Sivakumar,
P., Vinod, B., Devi, R. 2016).
The environment AUTOSAR API for agile technology platforms and adaptive users
at runtime is superior to the AUTOSAR classic platform.
Role of AUTOSAR in Automotive Software Trends 5
1.4.4 Mission of AUTOSAR
The AUTOSAR association was established for the purpose of standardizing a typi-
cal device, critical machine features, and useful interfaces. This allows advance-
ment allies to organize, exchange, reuse, and pass functions within an automobile
group, dramatically increasing their production performance. AUTOSAR moves the
change in perspective from just an ECU-based to a restriction contraption configura-
tion effort in vehicle programming improvement and makes the organization of the
constantly changing programming and E/E multifaceted nature for production and
financial aspects.
6 Software Engineering for Automotive Systems
1.4.5 Purpose of AUTOSAR
AUTOSAR is a global development collaboration of vehicle-interested parties that
was established in 2003. Its aim is to develop and implement a flexible and modular
software program framework in automobile computer-assisted control systems.
1.4.6 Methodology
The system configuration description includes all framework records as well as
information shared among various ECUs (for example, meaning of transport or
BUS signals). The ECU-specific statics contain the statics from the configuration
management description that are needed for an exact ECU (for instance, those
alarms in which a remarkable ECU is approaching). The ECU setup summary is
a file that contains all of the essential computer program setup information for a
specific ECU.
• Abstraction of hardware
• Runnable and job scheduling (OS) AUTOSAR
• Services for diagnosis and therapeutics
• Services for security and safety
growing and evolving standard that defines a layered architecture for software. The
AUTOSAR tier, for example, is separated into three layers and operates upon the
microcontroller. Let’s take a closer look at these:
1.6.2 Service Layer
The highest layer in the AUTOSAR core system design is the provider layer. The
provider layer creates a working system that operates from the product (S/W) layer to
a base microcontroller. The OS serves as a link between the microcontroller as well
as the application layer, and it can schedule utility tasks. Network services, storage
services, networking services, ECU country management, therapeutics services, and
other services are all handled by the carrier layer in BSW. The service layer sits on
top of the ECU abstraction layer, in the key neutral of the equipment, and is respon-
sible for providing the application with the required BSW execution. The RTE layer
separates the core software from the application software (Kienberger, J. et al 2014).
1.6.4 Application Layer
The AUTOSAR programming structure’s application layer facilitates the imple-
mentation of customized functionalities. The application layer is made up of client-
defined programming segments that communicate with sensors and actuators using
BSW. It also includes a variety of software components and applications that per-
form precise tasks according to instructions. The AUTOSAR programming layer
is made up of three parts: (1) application programming segments, (2) programming
segment ports, and (3) port interfaces. AUTOSAR ensures consistent interfaces for
utility layer programming program components, and the utility layer programming
program components assist in the development of basic applications to support auto-
mobile capabilities. Methodologies to use a virtual Function Bus for specific ports
promote communication between computing components (Dafang, W. et al 2010).
Furthermore, these ports make it easier for computing components and BSW to
communicate. The abovementioned structure of AUTOSAR is its traditional phase,
which reinforces ongoing assurance requirements and imperatives (Webers, W.,
Thörn, C., & Sandkuhl, K. 2008). The critical stage is suitable for supporting pur-
poses in the field of system administration and well-being because of the microcon-
troller, which allows ECUs to handle automobile sensors and devices.
1.7.2 Programming Language
Adaptive AUTOSAR relies upon on C 14 language standard. Usually, C language
is the top focal point to boost any vehicle application; however, the complexity of
the adaptive platform led to the adoption of C. Compared to C language, C offers a
higher mechanism for records encapsulation and helps massive and allotted systems
in a higher manner.
1.7.3 OS
AUTOSAR adaptive platform runs on POSIX-based (PSE51). POSIX stands for
“Portable Operating System Interface” through which one can access the OS.
Through PSE 51, about 300 APIs can be used, which allows higher portability.
POSIX-based totally (PSE51) additionally helps preemptive multitasking and allows
a dynamic variety of tasks.
The features and architecture mentioned above explain the technical background of
the AUTOSAR adaptive platform to support future use cases of automotive trends
mentioned below.
1.8.1 Autonomous Driving
To plan the points of independent driving, automobiles have to have sensible ECUs
that can deal with the giant quantity of data. Autonomous motors have to commu-
nicate with the changing surrounding environment, manipulate alerts from the hun-
dreds of in-vehicle sensors, and deal with the massive data that flows in the clever
ECUs. Intelligent ECUs require in-vehicle software to be up to date in the course of
a vehicle’s existence cycle due to evolving exterior structures and increased function-
ality. Autonomous automobiles of stage four and level five (partially or independent
vehicles) use tremendously complex algorithms, maps, and sensor fusion to feature
an adaptive platform that helps in enabling them all (Parekh, T. et al 2021).
the hot tendencies in the linked vehicle paradigm (Subburaj, S. D. R. et al 2021). V2X
systems require invulnerable communication with different cars and off-board sys-
tems, which might also be non-AUTOSAR systems. Next-gen motors will be related
to different vehicles, smartphones, traffic infrastructure, etc., and in-vehicle V2X
applications will be required to be up to date OTA; this is the place the AUTOSAR
adaptive platform assists.
1.8.3 Vehicle Electrification
The other rising vehicle tendencies like automobile electrification require a complete
system-level approach in design, as its single function can create an effect on other
necessary functions. The features such as EV charging, charger conversation, and
hybrid electric powered motors are related to functionalities of electric cars only
and wanted to be addressed through sensible ECUs distinctively than in gasoline-
powered vehicles. The electric powered automobiles require extraordinary units
of ECU software and options when it comes to V2V, V2X connectivity, physique
area controller, and OTA software program updates. The AUTOSAR is the
solely successful platform that can meet the incremental requirements of electric
automobiles in communication, connectivity, and other areas.
1.8.6 AUTOSAR-Incorporated Applications
AUTOSAR is common in the industry for the communication of automotive systems.
AUTOSAR specifications describe a layer of BSW that includes services that com-
municate with precise hardware but have a general application interface. EMCOS
AUTOSAR is the eMCOS-compliant AUTOSAR Classic Platform profile, a real-
time running computer (real-time systems or embedded systems with integrated
RTOs) that was once the first such product available on the market to provide flexible
support ranging from single-core to multi-core processors. A standardized interface
is an interface that is predefined in the C language by using the AUTOSAR specifica-
tion as an API. It is used in an ECU between BSW modules, between the RTE and the
operating machine, or between the RTE and the BSW module. AUTOSAR is distinct
from all other embedded programming paradigms. The reason for the distinction
is that in uniform layers, distinguishing the output from complete ECU is needed.
Besides, via the modules or software program components, this split operation often
determines common interfaces. This standardization has a number of advantages,
including scalability to individual devices, acknowledgment of practicable safety
requirements, integration of multiple providers’ functional fashions, and reusable
software components. Aside from that, AUTOSAR is public architecture. A certifi-
cate from the board is required to be used for business purposes. A developer cannot
sell software under the AUTOSAR brand without the certificate from the board.
1.8.7 Organization
• A key associate
• Strategic associate
• Exclusive partnership
• Associate partner
• Development partner
The founding partners BMW, Bosch, Continental, Daimler AG, Ford, General
Motors, PSA Peugeot Citroen, Toyota, and Volkswagen are among the attendees’ key
associates. These companies are in charge of the AUTOSAR improvement partner-
ship’s organization, administration, and management. The Executive Board estab-
lishes the standard method and roadmap within this core. The steering committee
oversees non-technical operations such as partner admission, public family mem-
bers, and contractual issues. For that reason, the Chairman and Deputy Chairman,
each designated for a year, represent the Steering Committee. The contact with the
outdoor environment is taken over by the AUTOSAR spokesperson. Strategic asso-
ciates are selected from the circle of premium partners for a two-year term and assist
the undertaking chief group in a variety of technological, operational, and day-to-
day processes. They also provide new strategic inputs to the project manager round.
Participants in the standard and community development contribute to work appli-
cations that are organized and supervised by the Project Manager Team, which was
formed with the help of the core partners. The popular files AUTOSAR has already
released are being used by associate partners. Attendees are actively working on
14 Software Engineering for Automotive Systems
1.9 ADVANTAGES
• Hardware and software are largely unrelated to one another.
• Horizontal layers will decouple development (through abstraction), reduc-
ing development time and costs.
• Software reusability improves consistency and performance.
• Create a development distribution system among suppliers.
• Improved design versatility allows you to compete on creative features.
• Make program and device integration easier.
• Lower the total cost of software creation.
• Enable more efficient variant handling.
• Share software modules between OEMs.
• Improve application development performance.
• Come up with new business models.
• Work in the planning process.
• Integrate software into a larger tool environment.
• Use normalized applications to enable new business opportunities.
• Gain a clear understanding of how automotive software is created.
1.10 APPLICATIONS
• In 2008, the first automobiles with AUTOSAR technology were introduced
to the market. Today, the brand is well-known, and AUTOSAR products
are successfully used in a variety of vehicle ventures. The large proportion
of the world’s automakers is AUTOSAR partners, and the significant pro-
portion of them are either using or planning to use AUTOSAR technology.
Auto manufacturing by AUTOSAR OEM associates accounts for roughly
80% of global demand.
• Because of the standardization, there would be more component reuse.
Expenses for skill enhancement can be spread out over a longer period of
time. Another issue that reduces costs is the interchangeability of supplier
solutions. As a result, there is more opposition.
• AUTOSAR emphasizes on technological advancements rather than busi-
ness models. In the marketplace, there will be some answers to this query.
1.11 CONCLUSION
AUTOSAR encourages a broad impact on the advancement of electronics automation.
It assembles electronics, OEMs, semiconductors, and a variety of automation compa-
nies to produce a variety of creative products. Elucidation suggested by AUTOSAR
methodology extensively reconciles with modern evolution since enhanced effective-
ness with considerably diminished time and money plunged provides the liberty to
OEMs to select software alternatives from multiple suppliers.
Role of AUTOSAR in Automotive Software Trends 15
REFERENCES
(Devi, R. S., Sivakumar, P., & Balaji, R. 2018) Devi, R. S., Sivakumar, P., & Balaji, R. 2018.
AUTOSAR Architecture Based Kernel Development for Automotive Application.
In International Conference on Intelligent Data Communication Technologies and
Internet of Things (pp. 911–919).
(Arts, T., Hughes, J., Norell, U. 2015) Arts, T., Hughes, J., Norell, U., & Svensson, H. 2015. Testing
AUTOSAR Software with QuickCheck. In 2015 IEEE Eighth International Conference
on Software Testing, Verification and Validation Workshops (ICSTW) (pp. 1–4).
(Kienberger, J., Minnerup, P., Kuntz, S 2014) Kienberger, J., Minnerup, P., Kuntz, S.,
& Bauer, B. 2014. Analysis and Validation of AUTOSAR Models. In 2014 2nd
International Conference on Model-Driven Engineering and Software Development
(MODELSWARD) (pp. 274–281).
(Parekh, T., Kumar, B. V., Maheswar, R. 2021) Parekh, T., Kumar, B. V., Maheswar, R.,
Sivakumar, P., Surendiran, B., & Aileni, R. M. (2021). Intelligent Transportation
System in Smart City: A SWOT Analysis. In Challenges and Solutions for Sustainable
Smart City Development (pp. 17–47). Springer, Cham.
(Mahmud, N., Rodriguez-Navas, G., Faragardi, H. 2018) Mahmud, N., Rodriguez-Navas,
G., Faragardi, H., Mubeen, S., & Seceleanu, C. 2018. Power-aware Allocation of
Fault-tolerant Multirate AUTOSAR Applications. In 2018 25th Asia-Pacific Software
Engineering Conference (APSEC) (pp. 199–208).
(Webers, W., Thörn, C., & Sandkuhl, K. 2008) Webers, W., Thörn, C., & Sandkuhl, K. 2008.
Connecting Feature Models and AUTOSAR: An Approach Supporting Requirements
Engineering in Automotive Industries. In International Working Conference on
Requirements Engineering: Foundation for Software Quality (pp. 95–108).
(Fürst, S., & Bechter, M. 2016) Fürst, S., & Bechter, M. 2016. AUTOSAR for Connected
and Autonomous Vehicles: The AUTOSAR Adaptive Platform. In 2016 46th annual
IEEE/IFIP International Conference on Dependable Systems and Networks Workshop
(DSN-W) (pp. 215–217).
(Subburaj, S. D. R., Kumar, V. V., Sivakumar, P. 2021) Subburaj, S. D. R., Kumar, V. V.,
Sivakumar, P., Kumar, B. V., Surendiran, B., & Lakshmi, A. N. 2021. Fog and Edge
Computing for Automotive Applications. In Challenges and Solutions for Sustainable
Smart City Development (pp. 1–15). Springer, Cham.
(Honekamp, U. 2009.) Honekamp, U. 2009. The Autosar XML Schema and Its Relevance for
AUTOSAR Tools. IEEE Software, 26(4), 73–76.
(Elbahnihy, A., Safar, M., & El-Kharashi, M. W. 2020) Elbahnihy, A., Safar, M., &
El-Kharashi, M. W. 2020. Hardware-accelerated SOME/IP-based Serialization for
AUTOSAR Platforms. In 2020 15th Design & Technology of Integrated Systems in
Nanoscale Era (DTIS) (pp. 1–2).
(Dafang, W., Jiuyang, Z., Guifan, Z. 2010) Dafang, W., Jiuyang, Z., Guifan, Z., Bo, H.,
& Shiqiang, L. 2010. Survey of the AUTOSAR Complex Drivers in the Field of
Automotive Electronics. In 2010 International Conference on Intelligent Computation
Technology and Automation (Vol. 3, pp. 662–664).
(Ryu, H., Jnag, S. Y., & Lee, W. J. 2013) Ryu, H., Jnag, S. Y., & Lee, W. J. 2013. AUTOSAR Unit
Testing Approach Based on Virtual Prototyping for Software Components. In 2013 Fifth
International Conference on Ubiquitous and Future Networks (ICUFN) (pp. 114–116).
(Sivakumar, P., Vinod, B., Devi, R. 2016) Sivakumar, P., Vinod, B., Devi, R. S., & Divya,
R. (2016). Deployment of Effective Testing Methodology in Automotive Software
Development. Circuits and Systems, 7(9), 2568–2577.
(Wu, R., Li, H., Yao, M. 2010) Wu, R., Li, H., Yao, M., Wang, J., & Yang, Y. 2010. A Hierarchical
Modeling Method for AUTOSAR Software Components. In 2010 2nd International
Conference on Computer Engineering and Technology (Vol. 4, pp. V4–184).
16 Software Engineering for Automotive Systems
(Singh, G., Kamath, N., & Sharma, R. K. 2018) Singh, G., Kamath, N., & Sharma, R. K. 2018.
Implementing Adaptive AUTOSAR Diagnostic Manager with Classic Diagnostics as
APIs. In 2018 Second International Conference on Intelligent Computing and Control
Systems (ICICCS) (pp. 894–898).
(Aust, S. 2018) Aust, S. 2018. Paving the Way for Connected Cars with Adaptive AUTOSAR
and AGL. In 2018 IEEE 43rd Conference on Local Computer Networks Workshops
(LCN Workshops) (pp. 53–58).
(Sivakumar, P., Vinod, B., Devi, R. S. 2016) Sivakumar, P., Vinod, B., Devi, R. S., & Divya,
R. 2016. Novelty Testing Measures and Defect Management in Automotive Software
Development. Australian Journal of Basic and Applied Sciences, 10(1), 607–613.
(Kim, J. W., Lee, K. J., & Ahn, H. S. 2015) Kim, J. W., Lee, K. J., & Ahn, H. S. 2015.
Development of Software Component Architecture for Motor-driven Power Steering
Control System Using AUTOSAR Methodology. In 2015 15th International Conference
on Control, Automation and Systems (ICCAS) (pp. 1995–1998).
(Kum, D., Park, G. M., Lee, S. 2008) Kum, D., Park, G. M., Lee, S., & Jung, W. 2008.
AUTOSAR Migration from Existing Automotive Software. In 2008 International
Conference on Control, Automation and Systems (pp. 558–562).
(Park, G., Kum, D., Jin, S. 2008) Park, G., Kum, D., Jin, S., & Jung, W. 2008. Implementation
of AUTOSAR I/O Driver Modules for a SSPS System. In 2008 International Conference
on Control, Automation and Systems (pp. 1451–1456).
(Swetha, S., & Sivakumar, P. 2021) Swetha, S., & Sivakumar, P. 2021. SSLA Based Traffic
Sign and Lane Detection for Autonomous cars. In 2021 International Conference on
Artificial Intelligence and Smart Systems (ICAIS) (pp. 766–771).
(Jayan, J., & Srinivasan, G. 2019) Jayan, J., & Srinivasan, G. 2019. AUTOSAR Based Dual
Core Partitioning for Power Train Application of BMS. In 2019 IEEE International
Conference on Intelligent Techniques in Control, Optimization and Signal Processing
(INCOS) (pp. 1–6).
(Sivakumar, P., Devi, R. S., Lakshmi, A 2020) Sivakumar, P., Devi, R. S., Lakshmi, A. N.,
VinothKumar, B., & Vinod, B. 2020. Automotive Grade Linux Software Architecture
for Automotive Infotainment System. In 2020 International Conference on Inventive
Computation Technologies (ICICT) (pp. 391–395).
(Soltani, M., & Knauss, E. 2015) Soltani, M., & Knauss, E. 2015. Challenges of Requirements
Engineering in AUTOSAR Ecosystems. In 2015 IEEE 23rd International Requirements
Engineering Conference (RE) (pp. 294–295).
(Dafang, W., Shiqiang, L., Bo, H. 2010) Dafang, W., Shiqiang, L., Bo, H., Guifan, Z., &
Jiuyang, Z. 2010. Communication Mechanisms on the Virtual Functional Bus of
AUTOSAR. In 2010 International Conference on Intelligent Computation Technology
and Automation (Vol. 1, pp. 982–985).
(Sandhya, D. R., Sivakumar, P., & Balaji, R. 2019) Sandhya Devi R.S., Sivakumar P., Balaji R.
2019. AUTOSAR Architecture Based Kernel Development for Automotive Application.
In: Hemanth J., Fernando X., Lafata P., Baig Z. (eds) International Conference on
Intelligent Data Communication Technologies and Internet of Things (ICICI) 2018.
ICICI 2018. Lecture Notes on Data Engineering and Communications Technologies,
vol 26. Springer, Cham.
2 Use of Communication
Protocols in Automotive
Software Development
Process
R. S. Sandhya Devi
Department of EEE, Kumaraguru College of Technology,
Coimbatore, India
P. Sivakumar
Department of EEE, PSG College of Technology,
Coimbatore, India
B. Vinoth Kumar
Department of IT, PSG College of Technology,
Coimbatore, India
A. D. Buvanesswaran
PSG College of Technology, Peelamedu, India
CONTENTS
2.1 Communication Protocol................................................................................. 18
2.1.1 Need for Communication Protocol in Automotive
Software Development......................................................................... 18
2.2 Automotive Systems........................................................................................ 19
2.2.1 History of Automotive Inventions....................................................... 19
2.3 Automotive Communication Requirements....................................................20
2.4 Typical Subsystem Modules............................................................................ 21
2.5 Automotive Communication Technologies..................................................... 22
2.5.1 Wired Technologies............................................................................. 22
2.5.2 Control Area Network (CAN)............................................................. 22
2.5.3 Local Interconnect Network (LIN).....................................................24
2.5.4 MOST..................................................................................................24
2.5.4.1 Function Blocks (F Blocks)..................................................24
2.5.5 Flexray.................................................................................................25
2.5.6 Byteflight.............................................................................................25
DOI: 10.1201/9781003269908-2 17
18 Software Engineering for Automotive Systems
used several actuators and sensors to replace the mechanical system. For the commu-
nication between the actuator and sensor components, we go for standard protocol.
If we take an automotive system, the requirement starts from the application we
use and criticality of that application. This decides how much importance we can
give for designing the network. If we divide the automotive system based on the
application and purpose, we can go with chassis, power train, body electronics, and
infotainment. These subsystems have different networking needs and this has to be
addressed with low latency. Another important thing is that, these networks have to
work easily together; there should be compatibility between them. Automotive sys-
tems are hard real-time systems, so error is not acceptable. And also, a car manufac-
tured today with the existing technologies should be flexible enough to update with
the new inventions, as a car runs for a minimum of ten years on road. Ten years is
an ample time for inventions. When an automotive system using minimum networks
to solve various requirements is appreciable as it reduces complexity and time. In
order to reduce this complexity, we have to go for a common networking topology.
These networks should support future plug-ins, less time requirement and support
“Network in Networks” (Muller, C., & Valle, M. 2011).
that, CAN became more popular, but CAN has used only two layers in OSI model,
so there were several advancements in CAN like Device Net, CAN Kingdom, and
CANopen. These protocols used higher layers of OSI model which gives easy main-
tenances and usability.
As the need pushed them to move form hydraulics and mechanical components to
electronic control, they were in need of high-speed bus system. After several rounds
of research and implementation in cars, they came with the system called “X-by-
wire” system. Reasons for moving from hydraulics to electronics are many, but the
most important one will be cost and maintenance of these systems and also advance-
ment needs completely new implementation. Another reason is recycling of heavy
hydraulic components are tedious.
Advanced driver assistance system (ADAS) assists the driver in driving and
parking functions. This is the base for autonomous driving field (Xu, Y.N. et al 2008)
(Swetha, S., & Sivakumar, P. 2021). This is done with the help of sensors and cameras
positioned at the various points in the vehicle. Some if its features include lane
departure warning, break assistance, emergency braking, etc. (Broy, J., & Muller-
Glaser, K. D. 2007). To be specific, ADAS needs proper communication only, from
the sensor to the system, from the system to the driver.
All the above-mentioned systems come under driving systems, which is needed
in a car to work. Another important system is on board diagnostics (OBD). OBD is
a self-diagnostic technology, where a car records the error and stores it as a specific
code related to the error. This we call it as diagnostic fault codes (DFC). These DFCs
are retrieved by the technician when the vehicle is taken for service. Reaching each
subsystem and debugging is a time consuming and tedious job for the technicians.
OBD also needs communication network from almost all the subsystems in the car,
to record the fault codes accurately.
I/O I/O
ECU ECU
device
device
FIGURE 2.1 Control area network. (National Instrument white paper. 2020)
To carry out error testing on each frame’s contents, the CAN specification con-
tains a cyclic redundancy code (CRC). Both node frames with errors and an error
frame may be transmitted to indicate the error in the network. The controller differ-
entiates between global and local errors, and incase more errors arise, corresponding
nodes stop sending errors or they get themselves disconnected (Wittmann, R., &
Zitterbart, M. 2000).
CAN Physical Layers: CAN has multiple distinct physical layers that can be
used. Some elements of the CAN network are defined by these physical layers, such
as transmission rate, signaling models, cable properties, baud rate, etc.
The physical layers that are widely used are discussed below:
Low-Speed/Fault-Tolerant CAN Hardware: Normally, CAN networks are often
designed with two wires, can communicate with devices at speeds up to 125 kbit/s,
and provide fault-tolerant transceivers. Other names include CAN B and ISO 11898-3
for low-speed/fault-tolerant CAN. Comfort devices include traditional low-speed/
fault-tolerant devices in a car.
High-Speed/FD CAN: The popular physical layer by far is the high-speed CAN.
With two wires, high-speed CAN networks are implemented and enable communi-
cation at transfer rates of up to 1 Mbit/s (Di Natale 2012). Other high-speed CAN
names include CAN C and 118982 ISO.
Software-Selectable CAN Hardware: With National Instruments CAN hard-
ware products, we can build CAN network either of the onboard transceivers (high-
speed, low-speed/fault-tolerant, or single-wire CAN) to use the software-selectable
CAN modules. Multiple transceiver modules will be a better solution for applications
that use a wide range of communication protocols (Johansson, K. H., Törngren, M.,
& Nielsen, L. 2005).
Single-Wire CAN Hardware: Communication speed can be up to 33.3 kbit/s
(88.3 kbit/s in high-speed mode) in single wire CAN communication networks.
Other names include SAE-J2411, CAN A, and GMLAN for single-wire CAN.
Working of CAN Communication: CAN networks do not have any master for
controlling the communication. Instead, each node can write CAN frame into the
24 Software Engineering for Automotive Systems
network, after checking whether the bus is busy. While writing into the network, no
address of the receiver is included in the frame. Alternatively, an arbitration ID is
included, which will be read by all the CAN nodes. These nodes will decide whether
to accept the frame, based on the arbitration ID (Sivakumar, P., Vinod, B., Devi,
R. S. 2015) (Paul et al 2016).
2.5.4 MOST
For distinct functions, a MOST network must have a range of masters. Masters may
be housed in the same unit.
Timing Master: Controls network timing and, therefore, synchronization
between devices. Network master sets up the network and assigns system addresses.
Connection master sets up synchronous channels of communication between devices.
Power master
Control power supplies
Power-up and
shut down handles
2.5.5 Flexray
BMW and Daimler-Chrysler after examination found that the existing technologies
will not support the automobile industry for future development in technologies espe-
cially X-by-wire systems (Mane, S. P., Sonavane, S. S., & Shingare, P. P. 2011). So, they
found that Flexray offers a solution for implementing X-by-wire systems by removing
some of the fieldbuses currently in use, thus, decreasing the overall network density
(Poledna S. 2001). Basically, today all car manufacturers joined this consortium, and
in mid-2004, the protocol specification was made public. In future automotive sys-
tems, FlexRay is used in ECU communication which need faster data exchange.
2.5.6 Byteflight
Byteflight was introduced in 1996 by BMW, and then further developed by BMW,
ELMOS, Infineon, Motorola, and Tyco EC. Safety-critical systems need more band-
width, which may now be satisfied by CAN but not in the future. So, Byteflight provides
a solution for this. ABS is one of the applications. When Byteflight was developed,
the key requirements were versatility, increased bandwidth than CAN. Byteflight sup-
ports network speeds of up to 10 MBps. For X-by-wire systems, Byteflight is a nomi-
nee. It has, however, been expanded to be part of the protocol of FlexRay.
2.5.7 Other Technologies
CAN, LIN, and Byteflight are most commonly applied in data exchange used for
chassis, power train, air bag, and electronics for physics and remedies, and diagnos-
tics. These are the first auto-rationale networking implementations in history, but
many innovations have been used over the years. Some were recurrent, while others
were eventually removed.
The Distributed Systems Interface, a master/slave network with verbal exchange
rates of up to 150 Kbps, used for security-related applications.
Safe-by-Wire is a master/slave network used for airbag control. Safe-by-wire has
elements taken from CAN and 150 Kbps conversation rates are supported. Since
CAN and LIN are no longer considered secure enough for airbag control, this verbal
exchange protocol was designed and developed by the Sound-by-Wire Consortium.
Motorola Interconnect (MI) In the experience that it is a simple low-fee master/
slave network specially designed for smart sensors, MI is comparable to LIN. However,
in today’s car systems, LIN is closed to being the world general.
Initially, the combination of computing devices with multimedia gadgets was the
position of multimedia and infotainment.
26 Software Engineering for Automotive Systems
FIGURE 2.2 Network infrastructure of Volvo XC90. (Nolte, T., Hansson, H., & Bello,
L. L. 2005)
Use of Communication Protocols in Automotive Software Development 27
TABLE 2.1
The Different ECU used in Volvo XC90
There are many cars which uses 100 ECUs in total. It is becoming more and more
difficult to incorporate these subsystems into the communication networks. Volvo
uses the Volcano definition in the case of the XC90. It is also possible to perform a
timings analysis of the method by using the Volcano instruments.
2.6.1 ZigBee
The new low-cost and low-power wireless PAN standard is ZigBee (IEEE 802.15.4)
designed for applications like sensor and controls. Given the number of proprietary
systems with low data rates designed to meet the above specifications, there were
no criteria that fulfilled them. In addition, the use of such devices has posed major
interoperability concerns that ZigBee technology addresses, offering solution for
sensor and control applications. In December 2004, the first ZigBees specification
for wireless data communications was ratified by the ZigBee Alliance (with over
28 Software Engineering for Automotive Systems
120 company members). ZigBee offers network speeds of up to 250 Kbps and is
intended to be widely used for control and monitoring purposes in the automotive
industry as a sensor network.
2.6.2 Bluetooth
Bluetooth (IEEE 802.15.1) officially offers up to 3 Mbps (Bluetoothv2.0) network
speeds. Bluetooth technology offers low-power, low-cost, short-range communication.
As a potential automotive wireless networking technology, the automotive envi-
ronment is also very enticing.
The Bluetooth Special Interest Group (SIG) founded the Car Working Group in
December 1999 in response to this interest. The hands-free profile was the first of
many requirements planned from the Car Working Group for the application stage
(Kovačević, J., Kaprocki, N., & Popović, A. 2019).
The Bluetooth SIG set out a three-year plan for potential bluetooth enhancements in
November 2004. Quality of Service, security, power consumption, multicast capabili-
ties, and privacy improvements are prioritized targets. It is expected that long-range
performance improvements will increase the range of bluetooth-enabled sensors.
2.6.3 Other Technologies
Wi-Fi stands for wireless fidelity and for every form of IEEE 802.11 network is the
general term. Wi-Fi is used, for example, by the Car2Car Consortium, a non-profit
organization initiated by European vehicle manufacturers, for inter-vehicle commu-
nications (Devi, R. S., Sivakumar, P., & Balaji, R. 2018) (Sivakumar P. et al 2020).
Advanced drive assistance to minimize the number of incidents, decentralized float-
ing car data to enhance local traffic flow and performance, and comfort and business
applications for user communications and information services to drive and pas-
sengers are applications here. For example, the European network-on-wheels (NoW)
project is a research project working in this field.
UWB (IEEE802.15.3a), or ultrawide band, has become rival to the IEEE 802.11
standards. UWB supports high bandwidth communication. Collision detection sys-
tems and suspension systems reacting to road conditions are other possible vehicle
applications that could be assisted by UWB. However, as UWB is a young technol-
ogy, these applications are not yet accessible.
2.7 CONCLUSION
We have discussed about the trends in technologies of automotive communication,
the recognition of typical requirements and novel vehicle technology specifications
and addressing to what extent networking is available and upcoming technologies
are capable of satisfying such demands. The next steps in automotive communica-
tions were discussed, focusing on X-by-wire systems and applications for wireless
communications. Today, one of the greater problems is interconnecting potentially
modern automotive architecture networks that are heterogeneous. This will be
answered by implementing standardized middleware technologies.
Use of Communication Protocols in Automotive Software Development 29
REFERENCES
(Broy, J., & Muller-Glaser, K. D. 2007) Broy, J., & Muller-Glaser, K. D. 2007. The Impact
of Time-Triggered Communication in Automotive Embedded Systems. In 2007
International Symposium on Industrial Embedded Systems (pp. 353–356).
(Chávez, M. L. 2006) Chávez, M. L. 2006. Fieldbus Systems and Their Applications 2005.
In Proceedings Volume from the 6th IFAC International Conference.
(Devi, R. S., Sivakumar, P., & Balaji, R. 2018) Devi, R. S., Sivakumar, P., & Balaji, R. 2018.
AUTOSAR Architecture Based Kernel Development for Automotive Application.
In International Conference on Intelligent Data Communication Technologies and
Internet of Things (pp. 911–919).
(Devi, R. S., Sivakumar, P., & Sukanya, M. 2018) Devi, R. S., Sivakumar, P., & Sukanya,
M. (2018). Offline analysis of sensor can protocol logs without can/vector tool usage.
International Journal of Innovative Technology and Exploring Engineering, 8(2S2),
pp. 225–229
(Di Natale, M., Zeng, H., Giusto, P 2012) Di Natale, M., Zeng, H., Giusto, P., & Ghosal, A.
2012. Understanding and Using the Controller Area Network Communication Protocol:
Theory and Practice. Springer Science & Business Media.
(Johansson, K. H., Törngren, M., & Nielsen, L. 2005) Johansson, K. H., Törngren, M., &
Nielsen, L. 2005. Vehicle applications of controller area network. In Handbook of
Networked and Embedded Control Systems (pp. 741–765). Birkhäuser, Boston.
(Kovačević, J., Kaprocki, N., & Popović, A. 2019) Kovačević, J., Kaprocki, N., & Popović,
A. 2019. Review of Automotive Audio Technologies: Immersive Audio Case Study.
In 2019 Zooming Innovation in Consumer Technologies Conference (ZINC)
(pp. 98–99).
(Liebel, G., Tichy, M., Knauss, E. 2018) Liebel, G., Tichy, M., Knauss, E., Oscar Ljungkrantz
& Gerald Stieglbauer. 2018. Organisation and communication problems in automotive
requirements engineering. Requirements Engineering, 23, 145–167.
(Mane, S. P., Sonavane, S. S., & Shingare, P. P. 2011) Mane, S. P., Sonavane, S. S., & Shingare,
P. P. 2011. A Review on Steer-by-Wire System using Flexray. In 2011 2nd International
Conference on Wireless Communication, Vehicular Technology, Information Theory
and Aerospace & Electronic Systems Technology (Wireless VITAE) (pp. 1–4).
(Muller, C., & Valle, M. 2011) Muller, C., & Valle, M. 2011. Design and Simulation of
Automotive Communication Networks: the Challenges. e & i Elektrotechnik und
Informationstechnik, 128(6), 228–233.
(Nolte, T., Hansson, H., & Bello, L. L. 2005) Nolte, T., Hansson, H., & Bello, L. L. 2005.
Automotive Communications-past, Current and Future. In 2005 IEEE Conference on
Emerging Technologies and Factory Automation, Vol. 1, pp. 8–992.
(Paul, A., Chilamkurti, N., Daniel, A. 2016) Paul, A., Chilamkurti, N., Daniel, A., &
Rho, S. 2016. Intelligent Vehicular Networks and Communications: Fundamentals,
Architectures and Solutions. Elsevier.
(Poledna, S., Ettlmayr, W., & Novak, M. 2001) Poledna, S., Ettlmayr, W., & Novak, M. 2001.
Communication bus for automotive applications. In Proceedings of the 27th European
Solid-State Circuits Conference (pp. 482–485).
(Rappl, M., Braun, P., von der Beeck, M. 2002) Rappl, M., Braun, P., von der Beeck, M.,
& Schröder, C. 2002. Automotive software development: A model-based approach
(No. 2002-01-0875). SAE Technical Paper.
(Robert Bosch Gmbh 2013) Robert Bosch Gmbh. 2013. Bosch Automotive Electrics and
Automotive Electronics. In Systems and Components, Networking and Hybrid Drive,
5th edition. Springer Vieweg.
(Sandhya, D. R., Sivakumar, P., & Balaji, R. 2019) Sandhya Devi R.S., Sivakumar P.,
Balaji R. 2019. AUTOSAR Architecture Based Kernel Development for Automotive
Application. In Hemanth J., Fernando X., Lafata P., Baig Z. (eds) International
30 Software Engineering for Automotive Systems
B. Vinoth kumar
Department of IT, PSG College of Technology,
Coimbatore, India
P. Sivakumar
Department of EEE, PSG College of Technology,
Coimbatore, India
A. Neeraja Lakshmi
Department of EEE, PSG College of Technology,
Coimbatore, India
R. Tripathy
Product Engineering Services, KPIT Technologies Ltd.,
Pune, India
CONTENTS
3.1 Introduction..................................................................................................... 32
3.2 Existing Systems.............................................................................................. 33
3.3 Proposed System: Unified Diagnostic Services (UDS)...................................34
3.4 Flash Bootloader.............................................................................................. 35
3.5 UDS-Based Bootloader................................................................................... 36
3.6 System Implementation: Basic Bootloader Stack............................................ 36
3.7 Application Software Flashing Using Bootloader........................................... 39
3.7.1 Hardware Requirements...................................................................... 39
3.8 Simulation........................................................................................................40
3.9 Secure ECU Software Update: Software Integrity.......................................... 41
DOI: 10.1201/9781003269908-3 31
32 Software Engineering for Automotive Systems
3.1 INTRODUCTION
Each electronic control unit (ECU) in an automobile is responsible for the execution
of a particular program. For instance, an ECU antilock braking system makes sure
brakes are not stuck during braking. The ABS software program in the ECU hard-
ware is capable of recording the vehicle speed as an input and is designed based on
this information to minimize the brake on the wheels.
Bootloader is the software algorithm which is performed during device boot.
Host of functionalities are provided by automotive ECUs (control units). Such
technologies and functionalities have become highly sophisticated and complex.
To the automotive original equipment manufacturers (OEMs) and suppliers, it has
become crucial to ensure that these software-driven control units are still working
in a secure and productive environment. This will only be done if the ECUs have
the current and patched version of the software and security updates inside the
car. Hence, the framework developed and ported on the MCU platform will also
be modified very regularly, either from a remote location or at the service station.
This task of initiating the ECU software upgrade was assigned to the bootloader
program, which occupies the ECU ROM.
When looking at a luxury vehicle, though, one can note that it has almost 100 mil-
lion lines of software code. But this is very huge compared to an aircraft, which has
only six million lines of code. Therefore, owing to these large amounts of electronic
code, the firmware upgrade of an automotive ECU is a repetitive activity.
A bootloader software is designed to automate this flash reprogramming process
and to handle the firmware upgrade. The bootloader program tests that the latest/
updated version of the ECU program is usable at the device boot-up. If yes, then
bootloader program installs and stores the newly installed firmware version before
booting the device. Post this, the device boot-up is executed and device now runs in
a fully protected environment on the latest version of the program.
At over 100 million lines of code in the total design, more than a conventional
operating system, autonomous cars can no longer be viewed as pure mechanical
devices. Increasingly, cars are linked and computer-like, with the potential to inter-
act with cell phones, provide car passengers with the current weather and traffic
alerts, and relay safety details to other automobiles and facilities surrounding them.
Vulnerabilities in vehicle communications result in four obstacles to vehicle safety
such as restricted access, restricted numerical capacity, volatile attack scenarios, and
threats and essential danger to the life of drivers or passengers.
The main objective of this chapter is to develop a bootloader software using uni-
fied diagnostic protocols. The technique has to be implemented using any of the com-
munication channel connecting the ECU. The bootloader should be able to provide
a secure ECU reprogramming every time whenever a software update is requested.
In order to mitigate the security treats faced by the connected vehicles, the software
Bootloader Design for Advanced Driver Assistance System 33
car and system vendors to reduce protection and service threats as software is applied
to automobile fleets, but there was no end-to-end open-source approach to handle
upgrades until recently. Advanced telematic systems (ATS) has been collaborating
with GENIVI and AGL in their development/reference networks to incorporate stable
app updates. The paper provides an analysis of current open source OTA solutions,
implemented the GENIVI SOTA open source solution, defined its design and security
features, and defined the incorporation of the OTA system into the GENIVI develop-
ment platform and the AGL framework.
Linux-based automotive security mechanisms could be used to make the connected
cars safer and more secure in the future. The idea of virtualization is explored to
establish the impression of several private subsystems or guest systems inside a single
host system (Foll and Bollo, 2016). Each virtualization framework has frameworks
for minimizing the usage of resources. Virtualization may be more or less complete,
based on the model selected. Every computer has in certain cases a private kernel,
with its own dedicated hardware resources. In other instances, virtual hosts share
nearly everything: a single kernel, file system, RAM, I/O device with very limited
separation, such as a private collection of libraries to curtail RAM or processor use.
Security hardening features based on ARM’s TrustZone technology for info-
tainment systems is proposed that ensures confidentiality and integrity of critical
applications (Maksym, 2018). In addition, a strategy is introduced that enables miti-
gating the effect of certain attacks on the internal network of the vehicle. TrustZone
is a hardware design, introduced with ARM processors on system-on-chip (SoC)
computers and offers a protection mechanism to combat several vulnerabilities in
embedded systems. TrustZone’s primary function is partitioning both hardware and
software resources into one of two worlds—safe, for protection subsystems and criti-
cal properties, and regular, for everything else. The planned research guarantees the
safe elements remain secure and are essential.
The APPSTACLE project is discussed that simplifies the production phase of
automotive systems by providing an accessible and safe software framework that
links a wide range of automobiles to the software through in-car and internet con-
nections (Pakanen et al., 2017). APPSTACLE project goal needs a wide spectrum
of skills from networking systems to software-based technologies and services. The
in-vehicle design tackles existing and lacking modules, utilities, protocols, APIs,
and layers of software needed for stable and safe incorporation of the Car2X in-car
framework. Direct cloud access standardization for in-car applications by a specific
Car2X gateway interface also introduces the need for the prototypical specific Car2X
gateway, which facilitates ECU wide remote access (e.g., through the gateway’s CAN
bus connectivity) to the cloud.
"Häh! Miksi?"
Hugh Ritson tyynnytti häntä selittämällä, että jos hän oli utelias
tietämään, mikä hänen äitinsä nimi oli, hän kyllä voi sen saada
selville ja lisäsi: "Mitäpä nimi silloin hyödyttää, jos teidän äitinne
kerran on kuollut?"
"Se on totta", sanoi Drayton, joka nyt oli tyyntynyt ja alkoi tulla
vakuutetuksi, että hänen vaivannäkönsä oli ollut turha.
"Saatte olla varma, että teidän isänne, missä hän nyt lieneekin, on
aika lurjus", sanoi Hugh Ritson.
"Ja mitäs sanoo siihen oikea aviomies — eikö hänkin ole sukkela
veitikka?"
Tuli oli kokonaan riittynyt. Iloton huone oli melkein pimeä ja ilma
sakea savusta ja tukahduttavan ummehtunut vanhasta tupakan ja
viinan hajusta. Drayton napitti pitkän takkinsa kurkkuun asti.
Sinä hetkenä onneton pikku raukka oli hyökännyt yli lattian Hughin
syliin, jossa vapisi kuin säikähtänyt lintu.
"Oi, minä tiesin, että tulisit — minä olin varma, että sinä tulisit",
hän sanoi kuivaten silmänsä, taas itkien ja kuivaten ne uudestaan ja
kohotti värisevät huulensa suudeltaviksi.
"Ei mikään — ei mikään nyt, kun sinä olet tullut. Sinä vain olit
tulematta niin kauan, niin hirveän kauan."
"Ah!"
"Ei se ole sitä! Minulla ei ollut ketään, joka olisi tuntenut sinut",
sanoi Mercy laskien äänensä kuiskaukseksi.
"Ei se ole minun silmäini vuoksi. Mutta sillä ei väliä. Oi, minä
tiesin, ettet sinä unohtaisi minua. Vain joskus iltaisin, kun hämärä
hiipi sisään ja minä istuin ihan yksin takan ääressä, minä ajattelin:
'Hän ei välitä minusta. Hän ei tule minua hakemaan.' Mutta se ei
ollut totta, eihän?"
"Sinä olet sairas, pikku raukka, mutta sinä tulet vielä onnelliseksi
— olethan sinä onnellinen nyt, etkös olekin?" sanoi Hugh.
"Ja nyt, Mercy", sanoi Hugh Ritson, "minä tahdon, että sinä olet
järkevä pikku nainen ja teet, kuten minä pyydän etkä puhu
sanaakaan. Teethän?"
"Kuuntele. Minä menen nyt ulos, mutta tulen pian takaisin. Minun
täytyy puhua rouva Draytonin kanssa, minullahan on jotakin
maksettavaa, tiedäthän."
"Hyvä on. Minä tapaan teitä vielä, rouva Drayton. Hyvää yötä,
Mercy, ja pidä kasvosi iloisina. Kas niin — suutele minua. Nyt, hyvää
yötä — mikä hullu, rakastava pikku hanhi sinä oletkaan — ja muista,
että olet vuoteessa ja nukut, ennenkuin minä palaan, muuten minä
julmasti suutun, niin aivan varmaan. Sinä et ole koskaan nähnyt
minua suuttuneena. No niin, älä siitä välitä. Hyvää yötä!"
"Minä en saa itkeä. Se tekee minun silmäni, oi, niin kipeiksi! Minun
täytyy hoitaa niitä hyvin ja saada ne terveiksi — niin, niin. Minun
täytyy olla terve ja vahva, kun — kun — se tapahtuu."
Kun Hugh Ritson astui ulos ja saapui tielle, oli yö pimeä. Tultuaan
juuri kapakan kellertävästä valosta hänen silmänsä tuskin erottivat
jalkakäytävän tai näkivät pensaitten epäselvän mustan rivin. Ilma oli
sumuinen. Kosteus keräytyi hänen kulmakarvoihinsa ja partaansa
suuriksi pisaroiksi, ja pakkanen jäädytti ne puikoiksi. Oli pureva
kylmä. Joesta nouseva höyry levisi kylmän viiman mukana yli
suoperäisen aukeaman, joka levisi oikealle ja vasemmalle tiestä.
Synkällä tiellä oli paksulti puoliksi jäätynyttä liejua.
"Seiso tässä, Greta", sanoi ääni, jonka hän hyvin tunsi. "Minä
palaan luoksesi pian. Anna minun koettaa auttaa noita tuolla."
"Minä pidin häntä, kunnes hänen takistaan jäi osa minun käteeni.
Katsokaa", sanoi toinen.
"Paul Ritson."
"Ja osoitteenne?"
"Hendon."
"Voin, herra."
Poliisikersantti liikahti.
"Palannut, herra? Ei, herra, on ihme, jos hän tulee kotiin ennen
aamua, herra; hän ei millään tavalla —"
"Lakatkaa lavertelemasta. Tyttö on kai omassa huoneessansa.
Menkää ja lukitkaa hänen ovensa."
"No, en, ehk'en. Mene nyt vain — kiireesti. Kuuletko? Miks'et sinä
mene?"
Hän lynkytti portaita alas niin pian kuin hänen kankeat raajansa
ikinä sallivat, mutta ovelta kuului koputus uudestaan.
"Minä olen täällä", vastasi Hugh astellen kapakan lattian yli kynttilä
kädessä oikealla-olevaan huoneeseen. Drayton seurasi häntä
koettaen nauraa. "Jouduinko ajoissa?"
"Peloitti kai."
"Juoksitte kai."
Drayton vaikeni.
"Te menette asemalle mainitun rouvan kanssa, kuten oli puhe.
Herra menee Lontooseen minun kanssani. He tulevat sittenkin tänne,
vaikka minun ensimmäinen tietoni oli väärä."
"Ei mitään kepposia, minä sanon sen teille. Jos te ette vie rouvaa
junaan — oikeaan junaan ja ole täällä takaisin huomenaamuna kello
puoli kaksi, te joudutte viettämään ihania päiviä Vanhassa
Baileyssä."
"Ei."
"Sen parempi."