MODULE - 4
Textbook 1:
Reema Thareja, Artificial Intelligence: Beyond Classical AI, Pearson Education,
2023
Chapter 8 (8.1, 8.2, 8.4), Chapter 9 (9.1- 9.3)
1. AI and Ethical Concerns
● Introduction to AI Ethics
● Ethical Implications in AI Development
● Addressing Bias and Fairness in AI
2. AI as a Service (AIaaS)
● Overview of AI as a Service
● Benefits and Challenges of AIaaS
● Popular AIaaS Platforms
3. Recent Trends in AI
● Overview of Current AI Trends
● Key Developments in AI Research and Applications
4. Expert Systems
● Introduction to Expert Systems
● Components of Expert Systems
● Applications of Expert Systems
5. Internet of Things (IoT)
● Introduction to IoT
● IoT Architecture and Components
● IoT Applications in Various Industries
6. Artificial Intelligence of Things (AIoT)
● AIoT: Combining AI and IoT
● Applications of AIoT in Smart Cities, Healthcare, and Industry 4.0
1. AI and Ethical Concerns
Ethics is about doing the right thing, even when no one is watching.
Examples:
● Being honest and truthful
● Respecting others
● Not harming people or society
● Acting responsibly in personal and professional life
1.1 Introduction to AI Ethics:
AI Ethics refers to the set of principles, guidelines, and moral values that guide the
development and use of Artificial Intelligence systems. It ensures that AI technologies are
created and used in a responsible, fair, and safe manner.
Definition
AI Ethics deals with understanding what is “right” or “wrong” when designing,
developing, and deploying AI systems. It aims to prevent misuse and ensure AI benefits society.
Need for AI Ethics
● AI is increasingly used in healthcare, finance, education, and daily life.
● Without ethical guidelines, AI systems may cause harm, bias, or violate privacy.
● Ensures trust, transparency, and accountability.
Major Ethical Principles
● Transparency: AI decisions must be explainable and understandable.
● Fairness (Avoiding Bias): AI should not discriminate based on gender, race, religion, etc.
● Privacy: AI systems must protect users’ personal data.
● Accountability: Developers and organizations are responsible for AI outcomes.
● Safety: AI should not cause physical, financial, or emotional harm.
Ethical Challenges in AI
● Data bias leading to unfair decisions
● Misuse of AI (e.g., deepfakes, surveillance)
● Lack of clarity in responsibility when AI fails
● Job displacement due to automation
Importance in Modern Society
● Builds trust in AI technologies
● Helps create regulations and standards
● Protects individuals and communities
● Ensures AI is used for societal well-being
1.2 Ethical Implications in AI Development
Ethical implications in AI development refer to the possible positive and negative effects
that AI systems can have on people, society, and the environment. Developers must consider
these issues to ensure AI is used safely and responsibly.
Bias and Discrimination
● AI systems can become biased if the training data is biased.
● This may lead to unfair decisions in areas like hiring, loans, or admissions.
Privacy Concerns
● AI collects and analyzes large amounts of personal data.
● If not handled properly, it can violate user privacy or lead to data misuse.
Lack of Transparency
● Many AI models work like a “black box.”
● If users cannot understand how decisions are made, it reduces trust and accountability.
Job Displacement
● Automation through AI can replace certain human jobs.
● This creates economic and social challenges for workers.
Accountability and Responsibility
● When an AI system makes a mistake, it is difficult to decide who is responsible—
the developer, the user, or the organization.
● Clear responsibility rules are needed.
Security Risks
● AI systems can be attacked or manipulated.
● If compromised, they may produce harmful outcomes.
Ethical Use of AI
● AI should not be used for harmful activities like surveillance misuse, deepfakes, or
spreading misinformation.
● Ethical guidelines ensure AI applications remain safe.
1.3 Addressing Bias and Fairness in AI
Addressing bias and ensuring fairness in AI is essential for creating reliable, trustworthy,
and ethical [Link] helps prevent discrimination and ensures AI benefits all sections of
society.
● Bias in AI occurs when an AI system makes unfair or unequal decisions due to biased
data or flawed design.
● Fairness means ensuring that AI treats all individuals equally without discrimination.
What is Bias in AI?
● Bias happens when the training data contains stereotypes or imbalanced information.
● This causes AI to favor or disadvantage certain groups (e.g., based on gender, age, or
region).
Why is Fairness Important?
● Unfair AI systems can harm people.
● Fairness builds trust, improves accuracy, and ensures ethical use of technology.
Sources of Bias
● Data Bias: Poor or unbalanced datasets.
● Algorithmic Bias: AI models that give unequal results.
● Human Bias: Developers’ assumptions influencing model design.
Methods to Address Bias
● Collect High-Quality Data: Use diverse and representative datasets.
● Data Cleaning: Remove incorrect, duplicated, or misleading data.
● Algorithm Testing: Regularly test AI for unfair outcomes.
● Use Fairness Metrics: Evaluate models using fairness standards.
● Human Oversight: Experts should review AI decisions.
Ensuring Fairness
● Promote transparency in how AI makes decisions.
● Follow ethical guidelines and industry standards.
● Continuously monitor AI systems after deployment.
1.4 Impact of AI on Jobs
AI and Job Creation
● AI will change the way people work, just like the Industrial Revolution did.
● It may remove around 75 million jobs but will also create about 133 million new jobs in
the coming years.
● So, overall, AI will create more jobs than it replaces.
Predicted Job Losses by 2030
● By 2030, 30% of current jobs could be automated.
● This may affect 400–800 million workers around the world.
● About 375 million people may have to learn new skills and switch to different types of
jobs.
AI’s Role in Job Automation
● Jobs with repetitive and predictable tasks are most likely to be [Link]:
○ Waiters (robot waiters)
○ Cashiers
○ Data entry work
● AI cannot replace jobs that require emotions, empathy, ethics, creativity, or complex
decision-making.
Jobs That Will Grow Because of AI
● Creative Jobs: Artists, designers, doctors, and scientists—AI will assist them, not replace
them.
● Management Jobs: Managers will still be needed because managing people requires
emotions and communication.
● Technology Jobs: Programmers, data scientists, AI engineers, and big data experts will be
in high demand.
AI’s Dual Impact
● AI removes low-skill, routine jobs but creates high-skill jobs that require programming,
robotics, and critical thinking.
● Future jobs will be more flexible, high-paying, and skill-based.
New Job Titles by 2030
AI will create completely new roles such as:
○ Chief Bias Officer
○ Data Detective
○ AI Tutor
○ AI Business Development Manager
AI’s Impact on Different Sectors
● Medical: Helps in diagnosis, predicting diseases, and improving healthcare.
● Automotive: Supports self-driving cars and smart manufacturing.
● Cybersecurity: Detects and prevents cyber threats.
● E-commerce: Used in recommendation systems, chatbots, and managing stock.
● HR: Helps in resume screening and talent search.
● Legal: Assists in document handling and case analysis.
2. AI as a Service (AIaaS)
What is AIaaS?
AI-as-a-Service means companies can use AI tools from cloud providers instead of
creating AI systems themselves.
AIaaS is growing fast because it is affordable, easy to use, and helps companies get AI
benefits without building everything from scratch. It also helps businesses make better
decisions using data.
2.1. Factors Triggering the Growth of AIaaS
● Cloud Platforms:Affordable cloud services help companies store and manage data easily.
● Improved Data Storage:Modern storage systems are cheaper, faster, and more reliable.
● Huge Data Generation:Devices like smartphones, smart TVs, and IoT sensors create
massive [Link] data can be used for analysis and better decision-making.
● Automation Tools:Automated data analysis tools help businesses understand data
quickly without manual work.
2.2. Growth of AIaaS
Why is AIaaS Growing?
● Cloud platforms make AI easily accessible to all companies.
● No need to hire expert AI engineers or build expensive systems.
Key Benefits of AIaaS
● Focus on Core Business:Companies don’t need to learn AI—they can focus on their main
work.
● Cost Reduction:No need to buy expensive hardware or software.
● Lower Investment Risk:Companies don’t have to spend huge money upfront.
● Quick Deployment:AI solutions can be used immediately with pre-built tools.
● Better Data Insights:AI analyzes trends and helps companies make smarter decisions.
● Scalability:Companies can start small and expand their AI usage whenever needed.
2.3. Challenges of AIaaS
● Security Issues:Sharing data with cloud vendors can create security risks.
● Dependence on Vendors:Companies rely on third-party services, which may cause
delays.
● Lack of Transparency:Companies often don’t know how the vendor’s AI system works
internally.
● Data Governance Rules:Some industries (like banking or healthcare) have strict data
rules that limit the use of third-party AI services.
2.4. Major Vendors of AIaaS
● Big Providers
○ Amazon Web Services (AWS):Offers machine learning APIs, chatbots, and tools.
○ Microsoft Azure:Provides cognitive services, bot services, and machine learning
tools.
○ Google Cloud Platform (GCP):Helps developers build and deploy ML models
easily.
● Other Providers
○ MonkeyLearn: No-code tools for text analysis.
○ IBM Watson: Tools for chatbots, virtual assistants, and text analytics.
○ Google Cloud ML: Helps data scientists train and deploy ML models.
2.5. Market Growth
● SaaS Market Value: Around $133 billion.
● AI Platform-as-a-Service (AIPaaS):Expected to grow from $11 billion (2023) to $88.5
billion (2025).
3 Robotics
3.1 Artificially Intelligent Robot
AI vs Robotics
● AI: Helps machines think and make decisions like humans.
● Robotics: Focuses on building machines that can perform physical tasks.
3.2 Artificially Intelligent Robots (AIRs)
AIRs combine both AI and [Link] can think + act using AI algorithms.
Examples:
● Warehouse robots: Use AI to find the best path to move.
● Drones: Automatically return home when battery is low.
● Self-driving cars: Detect obstacles and drive safely.
3.3 Types of Robots
1. Pre-Programmed Robots
● Do simple, repetitive tasks.
● Example: Robots in car manufacturing.
2. Humanoid Robots
● Look or act like humans.
● Example: Sophia robot.
3. Autonomous Robots
● Work on their own without human help.
● Examples: Roomba vacuum, medical assistant robots.
4. Teleoperated Robots
● Controlled by humans from a distance.
● Examples: Drones for landmine detection, underwater repair robots.
5. Augmenting Robots (VR Robots)
● Help improve human ability.
● Examples: Prosthetic limbs, exoskeleton suits.
3.4 Robots Based on Human Control
● Independent Robots
○ Work completely on their own.
○ Used in dangerous or boring [Link]: Bomb disposal robots, deep-sea
robots.
● Dependent Robots
○ Need human [Link]: Prosthetic limbs controlled by human signals.
● Chatbots
○ Software robots that talk to [Link] in customer service.
3.5 Components of a Robot
● Control System
○ The robot’s “brain” that controls all actions.
● Sensors
○ Help robots sense their environment.
○ Examples: Cameras, microphones, light sensors.
● Actuators
○ Parts that help robots move.
○ Example: Electric motors.
● Power Supply
○ Provides energy (battery, solar, hydraulic systems).
● End Effectors
○ Tools attached at the robot’s end to perform tasks.
○ Examples: Grippers, robotic hands.
3.6 AI Technologies Used in Robotics
● Computer Vision: Lets robots “see” and understand images.
● NLP: Helps robots understand and respond to human language.
● Edge Computing: Faster processing near the robot.
● CEP (Complex Event Processing): Handles multiple real-time events.
● Transfer Learning: Uses pre-trained models to save time.
● Reinforcement Learning: Robots learn by trial and error.
● Affective Computing: Gives robots emotional understanding.
● Mixed Reality: Combines real and virtual worlds for training robots.
3.7 Planning and Navigation
● Cognition
● Ability of a robot to make decisions to reach its goal.
● Path Planning
● Finding the best route from the start to the destination while avoiding obstacles.
3.7.1 Competencies for Planning
● Strategic Problem Solving
○ Planning the best path to reach a goal.
● Tactical Problem Solving
○ Avoiding obstacles and reacting to real-world changes.
3.7.2 Key Terms in Trajectory Planning
● Trajectory Planning: Moving from point A to B safely.
● Configuration Space: All possible positions a robot can take.
● Free Space: Areas where the robot can move without hitting anything.
3.7.3 Constraints in Trajectory Planning
● Holonomicity: Ability of a robot to move freely in all directions.
● Dynamic Environments: Obstacles may move, making planning harder.
3.7.4 Planning Algorithms
● Artificial Potential Field
○ Goal attracts the robot; obstacles push it away.
● Sampling-based Planning
○ Selects sample points and searches for a path.
● Grid-based Planning
○ Divides the area into a grid and finds a safe path.
● Reward-based Planning
○ Robot chooses actions that give maximum future reward (used in RL).
3.7.5 Visibility Graph
● Connects visible points in the environment to find the shortest path around obstacles.
3.7.6 Voronoi Diagram
● Creates paths that keep the robot as far as possible from obstacles, improving safety.
3.7.7 Cell Decomposition
● Exact Cell Decomposition
○ Breaks the area into clean cells and marks free space.
● Approximate Cell Decomposition
○ Uses fixed-size grids; faster but may miss small paths.
3.7.8 Potential Field Path Planning
● Uses attractive (goal) and repulsive (obstacle) forces to guide the robot like gravity.
4 Recent Trends in AI
4.1 Collaborative Systems
● Machine–Human Collaboration
● AI and humans work together by combining each other’s strengths.
Examples
● Foldit: Humans + AI work together to fold proteins (used in medicine).
● Chess: Amateurs with AI can defeat grandmasters.
● Business: Some companies use AI as part of their board decision-making.
● Military: Drones and humans work together for safer missions.
● Exoskeletons: Robot suits help soldiers lift heavy loads.
Challenges
● People may become too dependent on AI.
● Human skills may reduce over time.
● Some jobs may be replaced.
4.2 Machines Assisting Humans
● AI helps humans improve their abilities in thinking, creativity, and communication.
Examples
● SEB Bank AI Assistant:Helps customers with banking tasks and understands their tone of
voice.
● In Healthcare: AI helps diagnose diseases faster.
● At Home: Smart assistants and smart devices help daily activities.
4.3 Algorithmic Game Theory
● This is the use of game theory + AI to help machines make smart and ethical decisions.
Types of Games
1. Cooperative vs Non-Cooperative:
* Cooperative → players form groups.
* Non-cooperative → each player works alone.
2. Symmetric vs Asymmetric:
* Symmetric → players have the same goals.
* Asymmetric → goals are different.
3. Perfect vs Imperfect Information:
* Perfect → all players can see all moves.
* Imperfect → some information is hidden.
4. Zero-sum vs Non-zero-sum:
* Zero-sum → one wins, one loses.
* Non-zero-sum → all players can benefit.
Key Concepts
● Nash Equilibrium:A situation where no player can do better by changing their strategy
alone.
● Inverse Game Theory:Creating games by understanding player strategies.
4.4 Multi-Agent Reinforcement Learning (MARL)
What is MARL?
● Many AI agents learn together by interacting with each other.
Example
● Self-driving cars:Cars learn to cooperate with each other to reduce traffic and avoid
accidents.
Challenges
● More agents = more complexity
● Harder to coordinate decisions.
4.5 Neuromorphic Computing
Definition
● AI systems designed to work like the human brain.
Features
● Very low power usage
● Faster parallel processing
● High flexibility
● Can adapt and handle faults easily
Applications
● Real-time systems
● Robots
● Energy-efficient AI systems
How It Works
● Uses Spiking Neural Networks (SNNs) that communicate like human neurons.
Examples
● Intel Loihi chip
● IBM TrueNorth chip
4.6 Challenges in Neuromorphic Computing
● Difficult to shift from traditional computers (Von Neumann systems).
● Requires new:
○ Programming languages
○ Sensors
○ Storage technologies
● Developers must learn new methods to write programs for neuromorphic systems.
4.7 Impact of AI Trends
● AI is now used everywhere (healthcare, business, education, transport).
● New job roles and technologies are emerging.
● But ethical challenges like bias, privacy, and misuse must be handled carefully.
5. Expert System
What is an Expert System?
● An Expert System is a computer program that acts like a human [Link] is used to solve
complex problems in areas like medicine, science, engineering, finance, etc.
● It helps humans in decision-making but does not replace human experts.
Key Components:
● User Interface: Allows interaction with the system.
● Knowledge Base (KB): Stores expert knowledge for decision-making.
● Inference Engine: Applies reasoning rules to extract knowledge and provide decisions.
How it Works:
● Expert systems analyze user queries.
● Extract knowledge from the knowledge base.
● Use reasoning and inference rules to generate results.
5.1 Popular Examples
1. DENDRAL
● Used in organic chemistry.
● Helps identify unknown molecules from mass spectrometry data.
2. MYCIN
● Medical expert system.
● Diagnoses bacterial infections and suggests antibiotics.
3. PXDES
● Detects the type and level of lung cancer using body scanning images.
4. CaDeT
● Helps in early cancer detection.
5. R1/XCON
● Used by companies to configure computers based on customer needs.
6. DXplain
● Medical support system that suggests possible diseases.
5.2 Characteristics of Expert Systems
● High Performance – Solves difficult problems accurately.
● Understandable – Gives answers in a simple, human-friendly format.
● Reliable – Works consistently without mistakes.
● Fast and Responsive – Gives quick solutions.
● No Memory Limit – Can store large amounts of information.
● Domain Expert – Works well only in the specific area it is designed for.
● Emotion-free – Not affected by stress or fatigue.
● Unbiased – Makes decisions purely on facts.
● Cost Effective – Cheaper than consulting human experts repeatedly.
● Non-perishable – Once created, it lasts for many years.
● Intelligent – Uses rules to solve problems like an expert.
5.3 Components of an Expert System
1. User Interface
* Allows users to ask questions and get answers.
* Accepts queries in simple language.
* Shows results in an understandable form.
2. Inference Engine (Brain of the System)
* Applies rules to the knowledge base.
* Draws conclusions and gives solutions.
* Works in two ways:
a. Forward Chaining
* Starts from facts → predicts what will happen next.
* Example: Market prediction.
b. Backward Chaining
* Starts from a goal → finds reasons behind it.
* Example: Disease diagnosis.
3. Knowledge Base
* Stores all expert knowledge.
* Two types:
* Factual knowledge (confirmed facts)
* Heuristic knowledge (experience-based rules)
* Usually stored in If-Then rules.
Case Study: MYCIN
1. Knowledge about bacterial infections is stored.
2. Doctor enters patient symptoms.
3. Inference engine applies rules.
4. System suggests probable disease + treatment.
5.4 Participants in Developing an Expert System
1. Expert
* Person who provides special knowledge (doctor, engineer, scientist).
2. Knowledge Engineer
* Collects expert knowledge and builds the system.
3. End User
* Person who uses the system to solve problems.
5.5 Capabilities of Expert Systems
● Give advice and recommendations.
● Help in decision-making.
● Solve complex problems.
● Diagnose problems (especially in medicine).
● Explain how decisions were made.
● Predict future outcomes.
● Suggest alternative solutions.
5.6 Advantages
● Gives accurate and consistent results.
● Works in dangerous environments safely.
● Easy to use and widely available.
● Saves time and reduces cost.
● Provides expert-level decisions anytime.
● Can explain the reason behind each decision.
5.7 Limitations
● Gives wrong results if knowledge is incorrect.
● Cannot think creatively or emotionally.
● Very expensive to build and maintain.
● Cannot learn from experience (no self-learning).
● Works only in a specific domain.
● Depends heavily on knowledge provided by experts.
● Needs frequent updates.
● Cannot handle unusual or rare cases.
● Follow the rule: Garbage In → Garbage Out (poor data = poor result).
5.8 Applications
● Design & Manufacturing: VLSI design, car design.
● Finance: Fraud detection, loan approval.
● Medical diagnosis: Disease detection.
● Planning & Scheduling: Airlines, cargo, project planning.
● Software development: Tracking progress.
● Monitoring systems: Fault detection in pipelines.
● Automobiles/Electronics: Vehicle fault diagnosis.
● Process Control: Controlling industrial processes.
5.9 Expert System Technology
● Hardware
○ Workstations, minicomputers, mainframes.
● Programming Languages
○ LISP and PROLOG are commonly used.
● Tools
○ Editors, debuggers, prototyping tools.
● Expert System Shells
○ Systems without a knowledge base.
● * Provide a ready structure for building expert systems.
○ Examples:
■ JESS (Java-based)
■ Vidwan
5.10 Development of an Expert System
● Step 1: Identify the Problem
○ Choose the problem area.
○ Check if an expert system can solve it.
● Step 2: Design the System
○ Select hardware, software tools, shells.
○ Decide how knowledge will be represented.
● Step 3: Develop a Prototype
○ Collect expert knowledge.
○ Represent knowledge using IF-THEN rules.
○ Add heuristic rules for intelligent reasoning.
● Step 4: Test and Refine
○ Test using real cases.
○ Fix errors and improve accuracy.
● Step 5: Complete the System
○ Integrate with databases and other systems.
○ Document the system.
○ Train users.
● Step 6: Maintain
○ Update the knowledge base regularly.
○ Add new features as needed.
6. Internet of Things (IoT)
Definition
● IoT = Network of connected devices (machines, people, animals, objects) that can collect
and share data automatically without human involvement.
● Each device has a unique ID and can communicate over the internet.
Examples
● People: Smart heart monitors
● Animals: Biochip tracking
● Cars: Tyre pressure sensors
● Objects: Any device with an IP address (smart bulbs, ACs)
6.1 Applications of IoT
● Precision Agriculture
○ Monitors soil, crops, weather → improves farming.
● Building Management
○ Automates lighting, heating, security.
● Healthcare
○ Wearables monitor patient health remotely.
● Energy
○ Smart grids, smart meters, power optimization.
● Transportation
○ Traffic control, vehicle monitoring, smart parking.
6.2 Examples of IoT Applications
● Smart Car Navigation
○ Suggests best route, avoids traffic, sends delay alerts.
● Smart Home Automation
○ Alarm rings → coffee machine starts automatically.
● Wearable Fitness Trackers
○ Smartwatches track steps, sleep, calories, productivity.
● Self-Driving Cars
○ Use sensors and AI to drive safely.
● Smart Football
○ Measures speed and distance for athlete training.
● Smart Refrigerator
○ Alerts when food expires or items are low.
● Smart Air Conditioner
○ Controlled from mobile before you reach home.
● Smart Cities
○ Reduce traffic, pollution, crime; better waste and resource management.
6.3 IoT Products
● Amazon Echo (Alexa)
○ Voice assistant that controls smart home devices.
● Fitbit One
○ Tracks steps, calories, sleep; syncs with phone/computer.
● Barcelona Smart City
○ Smart parking, environmental monitoring; leading global smart city model.
● Popular IoT Platforms
○ AWS
○ Microsoft Azure
○ IBM Watson
○ Cisco IoT
○ Oracle Cloud
○ Salesforce IoT
○ GE Predix
○ AT&T Connected Car
○ Millions of cars connected to the network for navigation, safety alerts, etc.
6.4 Challenges in IoT
1. Security Issues
* Devices can be hacked (cars, homes).
* Many devices lack encryption → unsafe data transfer.
2. Scalability
* Huge data requires cloud storage and big data analytics.
3. Interoperability
* Different devices use different standards → difficult to connect systems.
4. Lack of Standardization
* No universal rules → poor device compatibility.
5. Lack of Government Support
* No specific regulations for IoT safety and privacy.
6. IoT Malware & Ransomware
* Hackers can take control of devices and demand ransom.
7. Connectivity Issues
* Poor internet reduces IoT performance.
8. Privacy Concerns
* Data misuse (e.g., toys recording children’s voice).
9. Bandwidth Strain
* More devices → more data → high network load.
10. IoT Skills Gap
* Lack of trained professionals to manage IoT systems.
6.5 Sensors
Sensor Type Description Example
1. Oxygen Sensor - Detects the gasoline/oxygen ratio - Car emission control system
2. Motion Sensor - Detects interruption in energy flow - Home security lights, automatic
doors
3. Vision and Imaging Sensor - Detects objects or colors in view - Cameras for object detection
4. Temperature Sensor - Measures thermal parameters - Industrial processes for gases,
liquids, solids
5. Radiation Sensor - Detects alpha, beta, or gamma - Surveys and sample counting
6. Proximity Sensor - Detects presence of objects without contact - Manufacturing operations
7. Pressure Sensor - Detects force per unit area in gases/liquids - Control and display devices
8. Position Sensor - Senses position of objects like valves or doors - Control systems for
machinery
9. Photoelectric Sensor - Detects objects sense color - Manufacture material handling
automation
10. Particle Sensor - Detects airborne particulates - Bin and baghouse monitoring
11. Motion Sensor - Senses movement or stoppage - Conveyor detection, stalling bearings
12. Metal Detector - Senses the presence of metal - Security checks
13. Level Sensor Determines height of substances in tanks - Control and display systems
14. Leak Sensor - Detects unwanted discharge of liquids/gases - Monitoring gas leaks
15. Humidity Sensor - Measures water content in the air - Control systems for temperature
regulation
16. Gas and Chemical - Detects gases/chemicals - Environmental monitoring
17. Force Sensor - Measures forces like weight or torque - Control systems for load
measurement
18. Flow Sensor - Detects movement of gases/liquids/solids - Industrial processing systems
19. Flaw Sensor - Detects inconsistencies on surfaces - Manufacturing processes like welding
20. Flame Detector - Senses fire presence and quality - Combustion control systems
21. Electrical Sensor - Detects electrical parameters like current voltage - Monitoring
electrical systems
22. Contact Sensor - Detects physical touch or contact - Alarm systems for doors/windows
23. Non-contact Sensor - Does not require physical touch - Radar guns for speed detection
24. Speed Sensor - Measures speed of an object/vehicle - Vehicle monitoring
25. Ultrasonic Sensor - Measures distance using high-frequency sound waves - Object
distance measurement in robotics
7. Artificial Intelligence of Things (AIoT)
7.1 Introduction to AIoT
● AIoT = AI + IoT
● AI provides intelligence (learning, decision-making).
● IoT connects devices to collect and share data automatically.
● AIoT improves IoT by adding machine learning to make smarter decisions.
7.2 How AIoT Works
● AI is integrated into IoT devices.
● Devices collect data → AI analyzes it → generates insights.
● Edge AI processes data locally, reducing delay and network usage.
7.3 Where AI Enhances IoT
AI improves the Analysis stage of IoT by:
● Managing and analyzing large volumes of data.
● Giving fast and accurate results.
● Balancing cloud (central) and edge (local) intelligence.
● Ensuring data privacy and security.
7.4 Applications of AIoT
● Smart Cities: Traffic control, energy management.
● Smart Retail: Customer tracking, store optimization.
● Smart Homes: Personalized automation (lights/temperature).
● Manufacturing: Machine monitoring, fault detection.
● Autonomous Vehicles: Real-time decisions for driving.
● Healthcare: Patient monitoring and remote diagnosis.
● Fleet Management: Vehicle tracking, fuel monitoring.
7.5 Benefits of AIoT
● Higher operational efficiency.
● Real-time decision-making.
● Reduces workload through automation.
● Scalable and connects many devices.
● Better risk management (predict problems).
● Improves product and service quality.
7.6 Challenges of AIoT
● Security risks (sensitive data).
● Scalability issues (huge data volumes).
● Interoperability (devices from different vendors).
● Privacy concerns (user data protection).
7.7 Future of AIoT
● 5G will make AIoT faster and more powerful.
● Will revolutionize industries (manufacturing, healthcare, transport).
● More devices → more data → AI becomes essential to process it.
● Expected 42 billion IoT devices by 2025.
END OF MODULE - 4