Cyber Physical Systems

CSD3010
Unit - I

By
N.D. Patel
Email: narottamdaspatel@vitbhopal.ac.in
Linkden: https://www.linkedin.com/in/ndpatel18/
Contact Number: 9450095800
1
 Cyber Physical Systems
CSD3010
Unit - I

Introduction to Cyber Physical Systems: What are Cyber Physical

Systems? Memory Architectures- Axioms of Cyber Physical
Systems – infrastructure, hardware sensing/actuation, data
analysis, connectivity, and visualization. CPS vs IoT.

2
 Cyber Physical Systems
CSD3010
Unit - I

Modeling Cyber-Physical Systems: Overview of Continuous,

Discrete, and Hybrid Models. Sensors and Actuators, Continuous
Dynamics and Lyapunov Stability. Discrete Dynamics, Reactivity,
and Termination. Hybrid Dynamics. Composition.

3
Cyber Physical Systems
CSD3010

4
Network Security Fundamentals
 Introduction to Cyber Physical Systems:
• With the exponential growth of cyber-physical systems (CPS), new security
challenges have emerged. Various vulnerabilities, threats, attacks, and controls
have been introduced for the new generation of CPS.
Introduction to Cyber Physical Systems:
 Introduction to Cyber Physical Systems:

https://ieeexplore.ieee.org/abstract/document/7924372/?casa_token=-
MnrDaghpnAAAAAA:AmV4VilksYXz8jIlFZ03jkoWXZ2ua6lZ1chXn54Kq9zwLgFGPfT5gyMgi9o_FZ0ICagZ3AM
 Introduction to Cyber Physical Systems:

https://ieeexplore.ieee.org/abstract/document/7924372/?casa_token=-
MnrDaghpnAAAAAA:AmV4VilksYXz8jIlFZ03jkoWXZ2ua6lZ1chXn54Kq9zwLgFGPfT5gyMgi9o_FZ0ICagZ3AM
 2023 Emerging Technology Adoption Roadmap

https://www.gartner.com/en/articles/what-s-new-in-the-2023-gartner-hype-cycle-for-emerging-technologies
Cyber + Physical Systems
The term “cyber-physical systems” emerged in 2006, coined by Helen Gill at the National
Science Foundation in the US.

• It is combination of physics with cyber components networked which is interconnected.

• CPS-> monitor and control physical processes
• Sensors and Actuators
Cyber + Physical Systems-Applications
• Agriculture
• Aeronautics- improve aircraft control, improve performance, efficiency
• Healthcare- medical devices, wearables to monitors patients,
• Civil infra
• Manufacturing
• Transportation
NSF’s Definition of CPS
• Cyber-physical systems (CPS) are engineered systems that are built from,
and depend upon, the seamless integration of computation and physical
components.

• Advances in CPS will enable capability, adaptability, scalability, resiliency,

safety, security, and usability that will expand the horizons of these critical
systems.

• CPS technologies are transforming the way people interact with engineered
systems, just as the Internet has transformed the way people interact with
information.

• Is a system that integrates physical an computational components to monitor

and control the physical processes.
Application Domains – major societal impact
• Agriculture,
• Aeronautics,
• Building design,
• Civil infrastructure,
• Energy,
• Environmental quality,
• Smart city,
• Healthcare and personalized medicine,
• Manufacturing,
• transportation.
CPS: Application Domains – major societal impact
• Cyber + Physical
• Computation + Dynamics + Communication
• Security + Safety
Contradictions in CPS
• Adaptability vs. Repeatability
• High connectivity vs. Security and Privacy
• High performance vs. Low Energy
• Asynchrony vs. Coordination/Cooperation
• Scalability vs. Reliability and Predictability
• Laws and Regulations vs. Technical Possibilities
• Economies of scale (cloud) vs. Locality (fog)
• Open vs. Proprietary
• Algorithms vs. Dynamics
Challenges of Working in a Multidisciplinary Area
Challenges of Working in a Multidisciplinary Area

Small Computer

Connected Industrial
System

Network

BigComplex
System

Advanced
Manufacturing Robot
Automotive CPS
• Safer Transportation
• Reduced Emissions
• Smart Transportation
• Energy Efficiency
• Climate Change
• Human-Robot Collaboration
Example CPS System
• STARMAC Ǫuadrotor Aircraft

• Autonomous Rotocraft
 STARMAC Design Block

LIDAR: Light Detection and Ranging

GPS: Global Positioning System
NAVIC: Navigation with Indian Constellation
IMU: Inertial Measurement Unit
ESC: Electronic Speed Controller
UART: Universal Asynchronous Receiver-transmitter
What is this course about?
• A scientific structured approach to designing and
implementing embedded systems
• Not just hacking and implementing
• Focus on model-based system design, on embedded
hardware and software
Model, Design & Analysis
• Modeling is the process of gaining a deeper
understanding of a system through imitation.
Models specify what a system does.

• Design is the structured creation of artifacts. It

specifies how a system does what it does. This
includes optimization.

• Analysis is the process of gaining a deeper

understanding of a system through dissection. It
specifies why a system does what it does (or fails to
do what a model says it should do).
Model, Design & Analysis
 Project Ideas:

• https://www.raspberrypi.org/magpi/

• https://blog.adafruit.com/category/raspberry-pi/

• https://iccps.acm.org/2024/

• https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=ACM%2FIEEE+INTER
NATIONAL+CONFERENCE+ON+CYBER-PHYSICAL+SYSTEMS&btnG=
Introduction to Microcontrollers

26
Introduction to Microcontrollers

�A microcontroller (MCU) is a small

computer on a single integrated circuit
consisting of a relatively simple central
processing unit (CPU) combined with
peripheral devices such as memories, I/O
devices, and timers.
� By some accounts, more than half of all
CPUs sold worldwide are
microcontrollers.
� Such a claim is hard to substantiate
because the difference between
microcontrollers and general- purpose
processors is indistinct.

27
Microcontrollers

� An Embedded Computer System on a Chip

� A CPU
� Memory (Volatile and Non-Volatile)
� Timers
� I/O Devices
� Typically intended for limited energy usage
� Low power when operating plus sleep modes
� Where might you use a microcontroller?

28
What is Control?

� Sequencing operations
� Turning switches on and off
� Adjusting continuously (or at least finely) variable
quantities to influence a process

29
Microcontroller vs Microprocessor

� A microcontroller is a small computer on a single integrated

circuit containing a processor core, memory, and programmable
input/output peripherals.

� A microprocessor incorporates the functions of a computer’s

central processing unit (CPU) on a single integrated circuit.

30
Microcontroller vs Microprocessor

31
Types of Processors

� In general-purpose computing, the variety of instruction set

architectures today is limited, with the Intel x86 architecture
overwhelmingly dominating all.
� There is no such dominance in embedded computing. On the
contrary, the variety of processors can be daunting to a system
designer.
� Do you want same microprocessor for your watch, autonomous
vehicle, industrial sensor?

32
How to choose micro-processors/controllers?

� Things that matter

� Peripherals
� Concurrency & Timing
� Clock Rates
� Memory sizes (SRAM &
flash)
� Package sizes

33
Types of Microcontrollers

34
DSP Processors

� Processors designed specifically to support numerically

intensive signal processing applications are called DSP
processors, or DSPs (digital signal processors).
� Signal Processing Applications: interactive games; radar, sonar,
and LIDAR (light detection and ranging) imaging systems; video
analytics (the extraction of information from video, for example for
surveillance); driver-assist systems for cars; medical electronics;
and scientific instrumentation.

35
Multiply-Accumulate Instructions

� Digital Signal Processors provide a fast and efficient multiply-

accumulate (MAC) instruction
� Typically including a relatively large accumulator
� They also typically use a Harvard memory access architecture
� They may include auto-increment addressing modes
� They may support circular buffer addressing
� Efficient implementation of delay lines
� They may support zero-overhead loops

� An Accumulator store intermediate logical or arithmetic data in

multistep calculations.

36
Programmable Logic Controller (PLC)

� A microcontroller system for industrial automation

� Continuous operation
� Hostile environments
� originated as replacements for control circuits using electrical relays
to control machinery

� PLCs are frequently programmed using ladder logic

� This notation was developed to specify logic constructed with relays
and switches

� AI can use ML algo to improve PLC programming.

� To analze data from a PLC to identify patterns and optimize the control
parameters. 37
Ladder Logic & Relays

� Relay is a switch where the contact is

controlled by coil.
� When a voltage is applied to the coil, the
contact closes, enabling current to flow
through the relay.
� By interconnecting contacts and coils,
relays can be used to build digital
controllers that follow specified patterns. � Vertical Rails &
Horizontal
Is a graphical programming language. Rungs
� Contact: two
vertical bars
� Coil: circle
19
Example: explained

� Start/Run is a normally open contact

� Stop is normally closed, indicated by the slash
� It becomes open when the operator pushes the switch.
� When start is pushed, electricity flows
� Both Start and Run contacts close so that Motor runs
� When Start is released, Motor continues to run
� When Stop is pressed, current is interrupted and both Run contacts
become open and motor stops
� Contacts wired in parallel perform a logical OR function, and
contacts wired in series perform a logical AND.

39
GPUs

� A graphics processing unit (GPU) is a specialized processor

designed especially to per- form the calculations required in graphics
rendering.
� Most used for Gaming (earlier days)
� Common programming language: CUDA

� NVIDIA?
� NVIDIA GeForce RTX 4090
� NVIDIA GeForce GTX 660
� AMD Radeon RX 7900 XTX

40
Parallelism vs Concurrency

� Embedded computing applications typically do more than one thing

“at a time.”
� Tasks are said to be “concurrent” if they conceptually execute
simultaneously
� Tasks are said to be “parallel” if they physically execute
simultaneously
� Typically multiple servers at the same time

41
Imperative Language

� Non-concurrent programs specify a sequence of

instructions to execute.
� Imperative Language: expresses a computation as a
sequence of operations
� Example: C, Java
� How to write concurrent programs in imperative
language?
� Thread Library

42
Program Dependency – Sequential Consistency

� No dependency between
lines 3 and 4

� Line 4 is dependent on
Line 3

43
Thread Mapping on Processor

� OS Dependent Scheduler
� Static Mapping
� Basic Lowest Load (fill in Round Robin
fashion)
� Extended Lowest Load

44
Performance Improvement

� Various current architectures seek to improve performance by

finding and exploiting potentials for parallel execution
� This frequently improves processing throughput
� It does not always improve processing latency
� It frequently makes processing time less predictable
� Many embedded applications rely on results being
produced at predictable regular rates
� Embedded results must be available at the right time

45
Parallelism

� Temporal Parallelism – Pipelining

� Spatial Parallelism –
� Superscalar
� VLIW
� Multicore

46
RISC and CISC Architectures

� CISC – Complex Instruction Set Computer

�Multi-clock complex instructions
� RISC – Reduced Instruction Set Computer
� Simple instructions that can be executed within one
cycle

47
5 Cycles of RISC Instruction Set

� Instruction fetch cycle (IF)

� Fetch instruction from memory pointed by PC,
then increment PC
� Instruction decode/register fetch cycle (ID)
� Decode the instruction
� Execution/effective address cycle (EX)
� ALU operates on the operands
� Memory access (MEM)
� Load/Store instructions
� Write-back cycle (WB)
� Register-Register ALU instruction

48
Pipelining in RISC

data hazard (computed branch)

control hazard (conditional branch)

4 branch
Mu taken
Add

Zer
o?
Dec
ode

memor
data
Mu

Mu
memory

x
y
Instruc

P Regist

AL
bank

U
tion

C er

data hazard (memory read or ALU result)

fetch decode execute memory writeback
49
Simple RISC Pipeline

50
Pipelining Hazard

� Data Hazard
� Control Hazard
� Out-of-order Execution
� Speculative Execution

51
 Out-of-order Execution

hardware resources: hardware resources:

instruction memory A B C D E instruction memory A B C D
register bank read 1 A B C D E register bank read 1 A E
B C D

interlock
register bank read 2 A B C D E register bank read 2 A B E
C D
ALU A B C D E ALU A E
B C D
data memory A B C D E data memory A E
B C D
register bank write A B C D E register bank write A E C D E
B

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 11 12
cycle cycle

Reservation Table Reservation Tablewith

Interlocks
52
CISC

� DSPs are typically CISC machines

� Instructions support
� FIR filtering
� FFTs
� Viterbi decoding

53
CISC Instruction

� Texas Instruments TMS320c54x family of DSP processors

� Code
� RPT numberOfTaps - 1
� MAC *AR2+, *AR3+, A
� RPT: zero overhead loops
� MAC : Multiply accumulate
� a := a + x � y
� AR2, AR3 are registers
� A is the Accumulator

54
Multicore Architecture

� Combination of several processors in a single chip

� Real-time and Safety critical tasks can have dedicated
processors
� Heterogeneous multicore
� CPU and GPUs together

55
FPGAs

� Field Programmable Gate Arrays

� Set of logic gates and RAM blocks
� Reconfigurable / Programmable
� Precise timing

� System on Chip design

Zy 40
Fixed and Floating Point Numbers

� Programs may use float or double

� Many embedded processors do not have floating point
arithmetic hardware
� Conversion required, which makes it slow
� Imaginary Binary Point is considered for computation
� Binary point separates bits
� Decimal point separates digits
� Format x.y representation indicates
� x bits left & y bits right of binary point

57
Programmers need to guard

� Overflow – since higher order bits are discarded

� Truncation, if bits are chosen before operation
� Rounding – rounds to nearest full precision after
operation

58
History of ARM Processor

59
ARM Cortex Processors
� ARM Cortex-A family:
� Applications processors
� Support OS and high-performance
applications Such as
Smartphones, Smart TV
� ARM Cortex-R family:
� Real-time processors with high
performance and high reliability
� Support real-time processing and
mission-critical control
� ARM Cortex-M family:
� Microcontroller
� Cost-sensitive, support SoC

60
Raspberry Pi

� The Raspberry Pi 3 Model B+ is the latest product in Raspberry Pi

range.
� Broadcom BCM2837B0,
Cortex-A53 (ARMv8) 64-bit
SoC @ 1.4GHz
� 1GB LPDDR2 SDRAM
� 2.4GHz and 5GHz IEEE
802.11.b/g/n/ac wireless
LAN, Bluetooth 4.2,
BLE
� Gigabit Ethernet over USB
2.0 (maximum throughput
300 Mbps)
� Extended 40-pin GPIO header 61
 Raspberry Pi
� The Raspberry Pi 3 Model B+ is the latest product in Raspberry Pi range.
� CSI camera port for
connecting a Raspberry Pi
camera
� DSI display port for
connecting a Raspberry Pi
touchscreen display
� 4-pole stereo output and
composite video port
� Micro SD port for loading your
operating system and storing
data
� 5V/2.5A DC power input
� Power-over-Ethernet (PoE)
support (requires separate PoE 62
ARM Peripherals

63
Modeling Cyber-Physical Systems

CPS are Heterogeneous with both physical and cyber components.

64
 Modeling Cyber-Physical Systems

https://link.springe
r.com/chapter/10.
1007/978-3-319-
76935-6_5

65
Hybrid Cyber-Physical Systems

66
Cyber-Physical Systems in Smart
Grid

67
Modeling Cyber-Physical Systems: Example

68
Quantum Cyber-Physical Systems
00
01
10
11
Two qubits can represent
the exact same Four states

69
ML in Embedded and Cyber-Physical Systems

70
 ML in Embedded and Cyber-Physical Systems

https://aicps.eng.uci.edu/machine-learning-in-embedded-cyber-physical-systems/

71
 ML in Embedded and Cyber-Physical Systems

https://aicps.eng.uci.edu/machine-learning-in-embedded-cyber-physical-systems/

72
 Test bed for power side-channel data collection

https://aicps.eng.uci.edu/machine-learning-in-embedded-cyber-physical-systems/
73
 Test bed for power side-channel data collection

https://aicps.eng.uci.edu/machine-learning-in-embedded-cyber-physical-systems/
74
Cyber Physical System

Unit - III

By
N.D. Patel
Email: narottamdaspatel@vitbhopal.ac.in
Linkden: https://www.linkedin.com/in/ndpatel18/
Contact Number: 9450095800
75
 Cyber Physical System

Unit - III
Analyzing Cyber Physical Systems: Real-Time Tasks and Worst-Case
Execution Time. Finding and Proving Invariants. Convergence,
Liveness, and Termination using Lyapunov Functions, Multiple Lyapunov
Functions, Common Lyapunov Functions, and Ranking Functions.

76
 Analyzing Cyber Physical Systems
1) Physical
2) Networking and communication
3) Distributed Cyber System
Analyzing Cyber Physical Systems
Analyzing Cyber Physical Systems: How does CPS
Work?
CPS integrate sensing and communication function �
Networking into physical Objects
Infrastructure
Connecting all devices
Internet
Real-Time Tasks and Worst-Case Execution Time
(WCET)
Execution time and Distribution (Safety Margin/Over Margin)

Measured WCET
Exact WCET
Estimated WCET
WCET and Best-Case Execution Time (BCET)
Execution time and Distribution (Safety Margin/Over Margin)
 CPS: Finding and Proving Invariants
Cyber Invariant + Network Invariant + Physical Invariant = System Invariant

Network Invariant: IP, Mac, interface delay, propagation delay, end-to-end packet transmission time, bandwid

https://ieeexplore.ieee.org/document/6166223
CPS: Convergence-A CONNECTED OPERATING ENVIRONMENT

Analyze whether potential solutions meet your needs and criteria, and decide whether you can implement them
successfully.
CPS: Convergence-ORGANIZATIONAL CHALLENGE
CPS: Convergence- ORGANIZATIONAL SOLUTION
CPS: BENEFITS OF CONVERGENCE

For further information and resources on National Critical Functions: cisa.gov/national-critical-functions

CONVERGENCE CASE STUDIES
IoT Devices Impacted by Ripple20 Vulnerabilities, June 2020 – A group of 19
vulnerabilities known as Ripple20 impacted millions of connected devices, including
smart home devices, power grid equipment, healthcare systems, industrial gear,
transportation systems, mobile/satellite communications equipment, and commercial
aircraft devices. These high-risk vulnerabilities in critical IoT devices across
numerous sectors could lead to compromised data and device malfunctions. In
response, some vendors released their own recommendations to mitigate potential
risks. The Cybersecurity and Infrastructure Security Agency (CISA) issued an alert
on potential weaknesses that malicious actors could exploit and provided
recommended mitigation approaches, including performing an impact analysis and
risk assessment; minimizing network exposure for control system devices and
systems; and ensuring access points are updated to the latest software version.
Security researchers have identified additional impacted vendors and devices since
Ripple20 was initially discovered.
CPS
CPS-Sensors
Lyapunov Functions
• Lyapunov functions, titled after Aleksandr Lyapunov, are scalar functions that can
be used to verify the stability of equilibrium of an ordinary differential equation in
the concept of ordinary differential equations (ODEs)

• Lyapunov functions (also known as Lyapunov’s second method for stability) are
crucial in dynamical system stability and control theory.

• A concept comparable to Foster–Lyapunov functions is found in the theory of

general state-space Markov chains.

• An Autonomous Dynamical System

• It can be used to predict the future state of a system based on its current state.
Lyapunov Functions
A Lyapunov function is a scalar function established on phase space that can be used to show an
equilibrium point’s stability (behavior of the system dones’t change).
A state where the output of the control system remains CONSTANT, even when the input values vary.

Suppose V(X) be a continuously differentiable function in

the origin’s neighbourhood U. If the following requirements
are satisfied, the function V(X) is known as the Lyapunov
function for an autonomous system X’ = f(x).

1. V(X) >0 for all X � U\ {0}

2. (dV/dt) ≤ 0 for all X � U
3. V(0) = 0

To examine the stability of various differential equations

and systems, the Lyapunov function method is used. We
will limit ourselves to autonomous systems in the next
sections.
X’ = f(x) or

(dxi)/dt = fi (x1, x2, …xn)

Here, i = 1, 2, …n
Lyapunov Functions
A Lyapunov function is a scalar function established on phase space that can be used to show an
equilibrium point’s stability.

Assume we’re given a continuously differentiable function

V(X) = V(x1, x2, …, xn) in the origin’s neighbourhood U.
Letting V(X) be the value of all X � U\ {0} and V(0) in the
origin. These are functions of the kind, for example,

V(x1, x2) = ax12+bx22, V(x1, x2) = ax12+bx24, a, b>0

 Lyapunov Functions
A Lyapunov function is a scalar function established on phase space that can be used to show an
equilibrium point’s stability.

https://byjus.com/maths/lyapunov-functions/
Lyapunov Functions
A Lyapunov function is a scalar function established on phase space that can be used to show an
equilibrium point’s stability.
TORA (Translational Oscillator with Rotational Actuator)
system
• A translational oscillator with a rotational actuator (TORA) is an
underactuated mechanical system with two degrees of freedom.

• Specifically, the dynamic model of the TORA system is first given, and
the passivity property is analyzed.

• Then, taking the input saturation into account, an energy-based

Lyapunov function augmented with an auxiliary signal is constructed, and
an amplitude-limited control law is derived.

https://tora.soft112.com/
TORA (Translational Oscillator with Rotational Actuator)
system

https://tora.soft112.com/
Cyber Physical System

Unit - IV

By
N.D. Patel
Email: narottamdaspatel@vitbhopal.ac.in
Linkden: https://www.linkedin.com/in/ndpatel18/
Contact Number: 9450095800
97
 Cyber Physical System

Unit - IV
Data Analytics: Introduction to Machine Learning. Mathematical
Optimization - planning and guidance. Basics of Neural Networks.
Deep Learning.
Visualization Tools: Data Abstraction – Task Abstraction – Labels
for Validation – Visualization Framework and workflow design -
Augmented Reality. Cyber Twin Concept. Microsoft Cloud with
Power Business Intelligence.
98
 100 Terms for DATA SCIENTEST
A- Accuracy, AUC, ARIMA
B- Bias, Binomial Distribution, Bayes
C- Clustering, CNN, Confusion Matrix, Cross-Validation
D- DNN, Data mining, Decision Tree, Dimensionality Reduction
E- Ensemble, EDA, Entropy
F- F-Score, Feature Extraction, Feature Engineering
.
.
.
.Z- Z-test, Z-Sore,….etc
Introduction to Machine Learning
What is Data Science?

Machine Learning

Wikipedia describes Data Science

MACHINE
as: LEARNING
COMPUTER
MATHEMATICS SCIENCE

“a multi-disciplinary field that uses DATA

SCIENCE
scientific methods, processes, DATA

algorithms and systems to

STATISTICAL
PROCESSING
RESEARCH

extract knowledge and insights DOMAIN

EXPERTISE
from structured and unstructured
data.”
https://en.wikipedia.org/wiki/Data_science
What is ML?

“Programming computers to learn from experience should

eventually eliminate the need for much of this detailed
programming effort”
Arthur Samuel (1959) – Computer Scientist
Data
Classical Programming (Rules, if/else, etc.) Answer
Rules s
Data Trained ML Models Answer
ML Algorithms
Answers (Rules) s

New Similar
Data
Machine Learning Lifecycle

Answe
r

New Data/Re-
training Deployment
Business Data
Problem Processing Yes
• Data Collection
• Data
Preprocessing Model
ML Problem Meets
• Data Visualization Algorith • Training
Formulation Business
• Data Augmentation m • Tuning Goal?
• Feature • Evaluation
Engineering
• Etc.
No
Some Important ML Terms

ML Statistics/Math/other Simply Put

Label/Target/y Dependent/Response/Output Variable The thing you’re trying to predict

Data that help you make
Feature/x Independent/Explanatory/Input Variable
predictions
Feature Engineering Transformation Reshaping data to get more value
1d, 2d,… nd Dimensionality Number of features
A set of numbers embedded in a
Model Parameters Weights
model that can predict the labels
Finding the ‘best’ set of model
Model Training Optimization
parameters
Why ML? Why now?

Data
• larger amounts of data, easy to produce, collect and store
Compute
• powerful processing units, hardware acceleration, parallelization
Algorithms
• ML frameworks, libraries, improved and more efficient techniques

Data Trained ML Models Answer

ML Algorithms
Answers (Rules) s

New Similar
Data
Introduction to Machine Learning

DATA + ALGORITHMS/STATISTICAL +
LEARNING/ANALYSE/Recognize
Patterns/Predictions/Decisions

Create Inferences/Improving
Accuracy/Performance over time

A MACHINE TO IMITATE INTELLIGENT HUMAN

BEHAVIOR
Introduction to Machine Learning: Applications

• Image and Speech Recognition: Google search, etc

• NLP- Natural Language Processing: language
translation
• Recommendation Systems
• Medical Diagnosis
• Financial Forecasting
• Autonomous Vehicles
• Image Processing

• AutoML
• ……
Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Ranking algorithm within Amazon Search
Ranking Helping users find the most relevant thing

Giving users the thing they may be most

Recommendation interested in

Figuring out what kind of thing something

Classification is

Regression Predicting a numerical value of a thing

Clustering Putting similar things together

Anomaly Detection Finding uncommon things

Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Recommendations across the website
Ranking Helping users find the most relevant thing

Giving users the thing they may be most

Recommendation interested in

Figuring out what kind of thing something

Classification is

Regression Predicting a numerical value of a thing Amazon’s Choice

Clustering Putting similar things together

Anomaly Detection Finding uncommon things

Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Product classification for our catalog
Ranking Helping users find the most relevant thing

Giving users the thing they may be most

Recommendation interested in

Figuring out what kind of thing something

Classification is
High-Low Dress Straight Dress

Regression Predicting a numerical value of a thing

Clustering Putting similar things together

Anomaly Detection Finding uncommon things Striped Skirt Graphic Shirt

Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Predicting sales for specific ASINs
Ranking Helping users find the most relevant thing

Giving users the thing they may be most

Recommendation interested in

Figuring out what kind of thing something

Classification is

Regression Predicting a numerical value of a thing

Clustering Putting similar things together

Anomaly Detection Finding uncommon things Seasonality | Out of stock | Promotions

Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Close-matching for near-duplicates
Ranking Helping users find the most relevant thing

Giving users the thing they may be most

Recommendation interested in

Figuring out what kind of thing something

Classification is

Regression Predicting a numerical value of a thing

Clustering Putting similar things together

Anomaly Detection Finding uncommon things

Introduction to Machine Learning: Applications

Business/ML Description Example

Problem Fruit freshness
Ranking Helping users find the most relevant thing

Giving users the thing they may be most Before After

Recommendation interested in

Figuring out what kind of thing something

Classification is

Regression Predicting a numerical value of a thing

Clustering Putting similar things together

Good
Damage
Serious Damage
Anomaly Detection Finding uncommon things Decay
 Quick Development Toolkit for Augmented Reality
Visualization

https://www.mdpi.com/2076-3417/12/16/8338
Using Augmented Reality in Different BIM
Workflows
 Virtual and Augmented Reality Tool Workflow

https://www.researchgate.net/publication/257745453_A_
Cyber Physical System

Unit - II

By
N.D. Patel
Email: narottamdaspatel@vitbhopal.ac.in
Linkden: https://www.linkedin.com/in/ndpatel18/
Contact Number: 9450095800
117
 Cyber Physical System

Unit - II
Design of Cyber Physical Systems: Architectural Choices. Real-Time
Operating Systems.
Networking Embedded Systems. Sensors and Actuators. End-to-End
System Design.

118
Design of Cyber Physical Systems
5 levels cyber physical system architecture
Current status and advancement of cyber-physical
systems in manufacturing

https://www.sciencedirect.com/science/article/pii/S0278612515000400
 CPS – a Concept Map

https://ptolemy.berkeley.edu/projects/cps/
 Robust Cyber–Physical Systems

https://www.sciencedirect.com/science/article/pii/S0167739X15002071
 Robust Cyber–Physical Systems

https://www.sciencedirect.com/science/article/pii/S0167739X15002071
 Design of Cyber Physical Systems: Real-Time Operating
Systems
Cyber-physical systems are systems comprising both a physical part and a software part, whereby the physical
part of the system sends information about itself to the software part, and the software sends information, usually
in the form of commands, to the physical part.

https://www.mdpi.com/2571-8800/4/2/12
Difference between Cyber-Physical System(CPS) and Embedded
System
Key Features of CPS Key Features of Embedded System
Single-purpose
Sensing and Actuation Low-power
Networked Real-time processing
Real-time processing Small form factor
Safety and Security Custom hardware
 Cyber Physical Systems: End-to-End System Design

https://ieeexplore.ieee.org/document/6862654
Cyber Physical Systems: End-to-End System Design
 A smart factory CPS (SF-CPS) based on the CPS 8C
architecture

https://journals.sagepub.com/doi/pdf/10.1177/1687814018784192
 Cyber Physical Systems: Sensors and Actuators

https://www.sciencedirect.com/science/article/pii/S209580991830612X?via%3Dihub