S2 - BCSE429L – Cyber Physical

System Design (Lecture)

Dr.D.Subitha

SCOPE, VIT Chennai

(Ph : 9884210662, Email : [email protected])

 Agenda
• 5C Architecture
• Technology Platforms for CPS
• Abstraction Layers in Computing
• Static Vs. Dynamic Systems
5C Architecture
5C Architecture for Manufacturing
Company
5C Architecture
Levels of 5C Architecture
The flow of data and information in a CPS-enabled factory with machine

tools in the production line based on 5C CPS architecture .

Implementation of 5C CPS architecture
for factories
 Stages of 5C – Industry 4.0

• At the component stage, once the sensory data from critical components
has been converted into information, a cyber-twin of each component will
be responsible for capturing time machine records and synthesizing future
steps to provide self-awareness and self-prediction.

• At the next stage, more advanced machine data, e.g. controller parameters
would be aggregated to the component’s information to monitor the status
and generate the cyber-twin of each particular machine.

• These machine twins in CPS provide the additional self-comparison

capability. Further at the third stage (production system), aggregated
knowledge from components and machine-level information provides self-
configurability and self-maintainability to the factory.
• This level of knowledge not only guarantees a worry-free and near-zero
downtime production but also provides optimized production planning and
inventory management plans for factory management
 Technology Platforms in CPS
• Significance of CPS in various domains (e.g., healthcare,
transportation, manufacturing).
• CPS enhances efficiency, automation, and real-time decision-
making.
• CPS technology has evolved from rudimentary control
systems to sophisticated integrated platforms.
• Early examples include basic feedback control mechanisms,
while modern CPS platforms incorporate advanced sensors,
actuators, and communication networks for seamless
integration.
 Key Characteristics of CPS
• Key characteristics include real-time processing for
immediate responses, integration of physical and
computational components for seamless interaction, and
adherence to communication standards to ensure
interoperability.
 Components of CPS Technology
Platforms
• Sensors and Actuators
• Sensors collect data from the physical environment, while actuators
respond to that data by controlling or manipulating physical
processes. This interaction forms the foundation of CPS
functionality.
• Control Systems
• Control systems regulate and manage physical processes in CPS,
ensuring desired outcomes. Feedback loops allow continuous
adjustments based on real-time data, enhancing the system's
adaptability.
• Communication Networks
• Communication is crucial for sharing data between components in
CPS. Both wired and wireless communication technologies facilitate
real-time data exchange, ensuring seamless integration and timely
decision-making.
 Role of Computation Units
• Computational units, including Edge Computing and
Cloud Computing, process data in CPS.
• Edge Computing handles data locally for immediate
responses, while Cloud Computing manages larger-
scale data processing, optimizing overall system
performance.
 Connectivity Achieved in CPS
• Connectivity in CPS is achieved through a combination
of wired and wireless communication technologies.

• IoT protocols facilitate device communication, and

various network topologies, such as star or mesh, are
chosen based on system requirements.
 Platforms for Real-Time Processing

• Real-Time Operating Systems (RTOS) ensure the timely

execution of tasks in CPS, making them crucial for
applications that demand immediate responses.
• Examples include autonomous vehicles and medical
devices.
 Edge Computing Enhance CPS
• Edge Computing processes data locally,
reducing latency and improving response times
in CPS.
• It is beneficial for applications requiring quick
decisions.
• However, challenges include limited
computational resources at the edge .
 Cloud Computing in CPS

• Cloud Computing in CPS handles extensive data

processing tasks, providing scalability and storage
capabilities.

• Benefits include resource scalability, while challenges

involve potential latency and security concerns.

• Examples include smart cities and industrial applications.

 Security in CPS
• Security is paramount in CPS due to the potential impact
of cyber-attacks on physical processes.

• Encryption, authentication, and access control measures

are implemented to safeguard against unauthorized
access.

• Real-world examples underscore the importance of

robust security measures
 Security in CPS

• Successful CPS implementations include smart

grids, healthcare monitoring systems, and
autonomous vehicles.

• Failures highlight the importance of thorough

testing, security measures, and adaptability in
technology platforms.
 Future trends in Technology

• Future trends include the integration of AI for

enhanced decision-making.

• Blockchain for secure transactions, and 5G

networks for improved connectivity in CPS.

• These technologies will likely shape the

evolution of CPS technology platforms.
 Challenges and Opportunities

• Challenges include ethical considerations in CPS

development, interoperability issues, and the need for
standardized communication.

• Opportunities lie in addressing these challenges through

research and innovation, creating more robust and
adaptable CPS technology platforms.
 Abstraction Layers in Computing
• Computing systems are complex and involve various
components working together to perform tasks.

• An abstraction layer is a conceptual separation or

encapsulation of functionality, providing a specific interface
and hiding the underlying complexity of the system.

• Abstraction layers simplify the interaction between

components, allowing developers to work with higher-level
concepts without needing to understand the intricate
details of the lower-level implementations.
 Examples of Abstraction Layers in
Computing
• TCP/IP Protocol Stack:
– The Internet protocol stack is organized into layers (physical,
data link, network, transport, application), each abstracting away
specific functionalities.
• Java Virtual Machine (JVM):
– JVM acts as an abstraction layer between Java applications and
the underlying hardware, allowing platform-independent
execution.
• Graphics Processing Units (GPUs):
– GPU abstraction layers enable developers to interact with
graphics hardware without needing in-depth knowledge of GPU
architectures.
• Database Management Systems (DBMS):
– DBMS abstracts away the complexities of interacting with data
storage, providing a higher-level interface for applications.
 Static System & Dynamic System

• A static system is a system in which

output at any instant of time depends on
the input sample at the same time.

• A static system is a memoryless system.

 Dynamic System

• A dynamic system is a system in which

output at any instant of time depends on
the input sample at the same time as well
as at other times.
Consider the system. x(t) is the input and y(t) is the output
of the system.
y(t) = 2 x(t)
put t = 0, y(0) = 2 x(0)
put t = 1, y(1) = 2 x(1)
put t = −2, y(−2) = 2 x(−2)
In the above example, the output y(t) at ‘t’ instant depends
on the input x(t) at the same time instant ‘t’ (present input).
So this system is static.

• y(t) = x2(t)
•put t = 0, y(0) = x2(0)
•put t = 1, y(1) = x2(1)
•put t = −1, y(−1) = x2(−1)
• In the above example, the output y(t) at ‘t’ instant
depends on the input x(t) at the same time instant
‘t’ (present input). So this system is static.

• y(t) = t x(t)
• put t = 0, y(0) = 0 x(0) = 0
• put t = 1, y(1) = 1 .x(0) = x(1)
• put t = −1, y(−1) = −x(−1)
In the above example, the output y(t) at the ‘t’ instant
depends on the input x(t) at the same time instant
‘t’ (present input). So this system is static.

• y(t) = log [x(t)]

• put t = 0, y(0) = log [x(0)]
• put t = 1, y(1) = log [x(1)]
• put t = −1, y(−1) = log [x(−1)]
In the above example, the output y(t) at the ‘t’ instant
depends on the input x(t) at the same time instant
‘t’(present input). So this system is static.

• y(t) = 2 x(t) − 3 x(t−1)

• put t = 0, y(0) = 2 x(0) − 3 x(−1)
• put t = −2, y(−2) = 2 x(−2) − 3 x(−3)
• In the above example, the output y(t) depends on
present input as well as past input. So this system
is dynamic.

• y(t) = x(2t)
• put t = 0, y(0) = x(0)
• put t = 1, y(1) = x(2)
• put t = −1, y(−1) = x(−2)
• In the above example, the output y(t) depends on
present input, past input and future input. So this
system is dynamic.

• y(t) = x(t2)

• put t = 0, y(t) = x(0)

• put t = 1, y(t) = x(1)

• put t = −2, y(t) = x((−2)2) = x(4)

• In the above example, the output y(t)
depends on present input, past input and
future input. So this system is dynamic.
HOMOGENEOUS SYSTEMS
Heterogeneous cyber-physical
systems
 Architecture description
languages
• ADL enable exploration of programmable
architectures for a given set of application
programs under various design constraints such
as area, power and performance.
• The ADL is used to specify programmable
embedded systems, includingprocessor,
coprocessor and memory architectures.
• The ADL specification is used to generate a
variety of software tools and models facilitating
exploration and validation of candidate
architectures.