INTRODUCTION TO

AI AND APPLICATIONS
BETC105/205
Module 4
Vemana Institue of Technology
Department of CSE

Prepared by
Mr. Naveen H S, Mr. Gopala Krishna P, Ms. Kalaiselvi,
Assistant Professor, Assistant Professor, Assistant Professor,
Dept of CSE, VIT Dept of CSE, VIT Dept of CSE, VIT
Outline
Module 1: Introduction to Artificial Intelligence
1. Introduction to Artificial Intelligence 2. Machine Intelligence
1. Definition of Artificial Intelligence 1. Defining Intelligence
2. How Does AI Work? 2. Components of Intelligence
3. Advantages and Disadvantages of Artificial Intelligence 3. Differences Between Human and Machine Intelligence
4. History of Artificial Intelligence 4. Agent and Environment in AI
5. Types of Artificial Intelligence: 5. Search Algorithms:
a. Weak AI vs. Strong AI a. Uninformed Search Algorithms
b. Reactive Machines b. Informed Search Algorithms:
c. Limited Memory i. Pure Heuristic Search
d. Theory of Mind ii. Best-First Search Algorithm (Greedy Search)
e. Self-Awareness
6. Is Artificial Intelligence the Same as Augmented
Intelligence and Cognitive Computing?
7. Introduction to Machine Learning and Deep Learning
Outline
Module 2: Introduction to Prompt Engineering
1. Introduction to Prompt Engineering 3. Prompts for Creative Thinking
1. Overview of Prompt Engineering Introduction to Creative Thinking with Prompts
2. The Evolution of Prompt Engineering Unlocking Imagination and Innovation
3. Types of Prompts
4. How Does Prompt Engineering Work?
4. Prompts for Effective Writing
5. The Role of Prompt Engineering in Communication
Introduction to Writing with Prompts
6. The Advantages of Prompt Engineering
Igniting the Writing Process with Prompts
7. The Future of Large Language Model (LLM)
Communication

2. Prompt Engineering Techniques for ChatGPT

Introduction to Prompt Engineering Techniques
Instructions Prompt Technique
Zero, One, and Few Shot Prompting
Self-Consistency Prompt
Outline
Module 3: Machine Learning
Reema Thareja, Artificial Intelligence: Beyond Classical AI,Pearson Education, 2023.

1. Machine Learning in AI 4. Clustering Techniques

Overview of Machine Learning Techniques Introduction to Clustering
Introduction to Machine Learning Models Types of Clustering Algorithms

2. Regression Analysis in Machine Learning 5. Neural Networks

Basics of Regression Basics of Neural Networks
Linear and Non-Linear Regression Techniques Types and Applications of Neural Networks

3. Classification Techniques
Overview of Classification Algorithms
Naïve Bayes Classification
Support Vector Machine (SVM)
Outline
Module 4: Machine Learning
Reema Thareja, Artificial Intelligence: Beyond Classical AI,Pearson Education, 2023.

1. AI and Ethical Concerns 4. Expert Systems

Introduction to AI Ethics Introduction to Expert Systems
Ethical Implications in AI Development Components of Expert Systems
Addressing Bias and Fairness in AI Applications of Expert Systems

2. AI as a Service (AIaaS( 5. Internet of Things (IoT(

Overview of AI as a Service Introduction to IoT
Benefits and Challenges of AIaaS IoT Architecture and Components
Popular AIaaS Platforms IoT Applications in Various Industries

3. Recent Trends in AI 6. Artificial Intelligence of Things (AIoT(

Overview of Current AI Trends AIoT: Combining AI and IoT
Key Developments in AI Research and Applications Applications of AIoT in Smart Cities, Healthcare, and
Industry 4.0
 INTRO
,
BOS Chairman (CS/IS) - VTU Belagavi
Vice Principal and Dean - Academics
Canara Engineering College
 Module 4: Current Trends in Artificial Intelligence
Chapter 8 Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns 8.2 AI as a Service (AIaaS( 8.3 Robotics 8.4 Recent Trends in AI
8.1.1 Ethical Use of Artificial 8.2.1 Factors Triggering 8.3.1 Artificially Intelligent Robot 8.4.1 Collaborative Systems
Intelligence Growth of AIaaS 8.3.2 Characteristics of Robots 8.4.2 Machines Assisting Humans
8.1.2 Is AI Dangerous? Will 8.2.2 The Growth of AIaaS 8.3.3 Types of Robots 8.4.3 Algorithmic Game Theory and
Robots Take Over the World? 8.2.3 Challenges of AIaaS 8.3.4 Types of Robots Based on Computational Social Choice
8.1.3 Ethics in AI 8.2.4 Vendors of AIaaS Degree of Human Control 8.4.4 Multi-Agent Reinforcement
8.1.4 AI and Bias 8.3.5 Components of a Robot Learning (MARL)
8.1.5 Towards Ethical and 8.3.6 AI Technology Used in 8.4.5 Neuromorphic Computing
Trustworthy AI Robotics
8.1.6 Why is Ethical AI 8.3.7 Planning and Navigation
Important?
8.1.7 Impact of AI on Jobs
Module 4: Current Trends in Artificial Intelligence
Chapter 9 Where AI Is Heading Today?
9.1 Expert System 9.2 Internet of Things 9.3 Artificial Intelligence of Things
9.1.1 Popular Examples of the Expert System 9.2.1 Examples of Applications of IoT (AIoT(
9.1.2 Characteristics of an Expert System 9.2.2 IoT Products 9.3.1 How Does AIoT Work?
9.1.3 Components of an Expert System 9.2.3 Challenges 9.3.2 Where Does AI Unlock IoT?
9.1.4 Participants in the Development of 9.2.4 Sensors 9.3.3 Applications and Examples of AIoT
Expert System 9.3.4 Benefits and Challenges of AIoT
9.1.5 Capabilities of the Expert System 9.3.5 Future of AIoT
9.1.6 Advantages of Expert Systems
9.1.7 Limitations of Expert Systems
9.1.8 Applications of Expert Systems
9.1.9 Expert System Technology
9.1.10 Development of Expert Systems
Module 4: Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns
Understanding AI Ethics and Its Importance
AI ethics addresses the moral obligations and responsibilities of AI systems, their creators, and
their users. It involves ensuring that AI technologies align with human values and societal norms.
1. Roboethics:
2. Machine Ethics:
 Module 4: Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns
1. Roboethics:
Roboethics focuses on human responsibility when it comes to designing, building, using, and
interacting with artificial beings (like robots).
It covers a wide range of ethical questions regarding how humans should treat AI entities.
Some of the critical areas include:
Designing ethical robots: Ensuring that robots are built with features that align with
human morals, including their capabilities and limitations.
Treatment of robots: Addressing how humans should interact with AI entities, especially
those designed to resemble humans.

[Type here]
Module 4: Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns
2. Machine Ethics:
Machine ethics, on the other hand, concerns the moral behavior of AI systems or Artificial
Moral Agents (AMAs), especially those designed to interact with humans or make important
decisions.
These systems need to make moral decisions independently, without direct human
intervention. Examples of AMAs include self-driving cars or AI systems used in healthcare.
This includes:
Decision-making systems: AI algorithms that affect people’s lives directly, like automated
loan approvals, criminal sentencing, or hiring decisions.
Moral programming: Developing frameworks that allow AI systems to make ethical
decisions based on societal norms, laws, and ethics.
 Module 4: Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns
Key Ethical Concerns in AI
1. Can AI Make Human-like Decisions?
AI generates human-like outputs (e.g., loan approvals, healthcare decisions).
Do AI systems understand the consequences of their actions like humans do?
2. Are AI Decisions Truly Human-like?
AI decisions are based on data and probabilities.
But do they fully comprehend human context behind those decisions?
3. Justification of AI’s Decisions
Is the reasoning behind AI decisions justifiable?
Transparency is crucial for aligning AI decisions with ethical and legal frameworks.
4. Fairness in AI Decisions
Are AI decisions fair and free from bias?
AI systems should respect individual rights and avoid discrimination.
5. Trustworthiness of AI
Can we trust AI systems (e.g., self-driving cars) to make safe decisions?
AI must be reliable, transparent, and accountable to earn trust.
[Type here]
Module 4: Current Trends in Artificial Intelligence
8.1.1 Ethical Use of Artificial Intelligence
Bias-Free AI
AI systems should be free from bias, especially with complex algorithms like deep learning and GAN.
Explainability
AI decisions, especially in regulated industries, must be explainable (e.g., credit decisions in finance).
"Black box" AI systems that can't explain decisions are problematic.
Key Features of Responsible AI
a. Explainable: AI decisions should be understandable.
b. Monitorable: AI should be trackable and auditable.
c. Reproducible: AI systems must yield consistent results.
d. Secure: AI systems must be safe from attacks.
e. Human-Centered: AI should prioritize human well-being.
f. Unbiased: AI must avoid discrimination.
g. Justifiable: Decisions should be defendable and ethical.
Challenges in Regulation
Formulating laws for AI is difficult due to rapid advancements in technology.
Existing laws (e.g., privacy) may not cover new AI technologies like voice assistants (e.g., Alexa, Siri), which can be
exploited by malicious actors.
Module 4: Current Trends in Artificial Intelligence
8.1.2 Is AI Dangerous? Will Robots Take Over the World?
AI’s Potential for Harm
AI could be risky if specifically programmed for harm (e.g., autonomous weapons).
Autonomous drones and self-driving cars can become dangerous if misused or hacked.

Lack of Clear Accountability

Current laws don’t cover AI’s role in society, especially in terms of liability.
Questions arise like: Who is responsible for AI’s actions? Can AI systems be held accountable?

Job Automation Risk

AI might replace many jobs, especially low-skilled, repetitive tasks.
This could lead to unemployment for millions, particularly in industries like retail, hospitality,
and warehouse work.
Module 4: Current Trends in Artificial Intelligence
8.1.2 Is AI Dangerous? Will Robots Take Over the World?
Biased Algorithms
AI systems are vulnerable to bias if trained with biased data.
Biased AI can produce unfair results, impacting crucial decisions like hiring or loan approvals.

Privacy Concerns
AI systems collect large amounts of personal data, leading to privacy invasion.
Facial recognition technology can track individuals without consent, creating privacy risks.

Unreliable Information
AI can be used to create fake content (e.g., deepfakes, fake news).
Manipulated videos or images can spread misinformation or be used maliciously
Module 4: Current Trends in Artificial Intelligence
8.1.3 Ethics in AI
Ethical AI Definition
Ethical AI refers to AI systems designed to be fair, just, and unbiased.
It focuses on ensuring AI’s actions align with moral principles.
Importance of Ethical AI
Ethical AI ensures the technology benefits society while avoiding harm or unfair treatment.
It guarantees AI decisions are transparent and responsible.
Bias in AI
AI systems can produce biased results if trained on biased data.
Unchecked bias in AI can lead to unethical decisions, such as unfair hiring or loan approval.
Example of Unethical AI
Tay Chatbot: Microsoft’s Tay learned negative and harmful behaviors after interacting with trolls on Twitter.
The bot started spreading hate and offensive messages, forcing Microsoft to shut it down within 24 hours.
Features of Ethical AI
Uses unbiased data for training.
Follows the right learning model for the problem at hand.
Continuously monitored to ensure its outputs are right and fair.
Module 4: Current Trends in Artificial Intelligence
8.1.4 AI and Bias
What is AI Bias?
AI bias occurs when an algorithm produces results that are unfair or prejudiced towards a
particular group.
Example: If AI frequently misidentifies people of color in computer vision systems, it’s biased.

Examples of AI Bias
Doctors in Images: Search results for "doctor" mostly show white men.
Gender Bias in Professions: "Doctor" is often male, while "nurse" is female in many images and
videos.
Voice Assistants: Virtual assistants like Siri and Alexa use female voices, reinforcing gender bias.
Recognition Failures: Computer vision struggles with recognizing people of different skin colors
accurately.
Module 4: Current Trends in Artificial Intelligence
8.1.4 AI and Bias
What is AI Bias?
AI bias occurs when an algorithm produces results that are unfair or prejudiced towards a
particular group.
Example: If AI frequently misidentifies people of color in computer vision systems, it’s biased.

Examples of AI Bias
Doctors in Images: Search results for "doctor" mostly show white men.
Gender Bias in Professions: "Doctor" is often male, while "nurse" is female in many images and
videos.
Voice Assistants: Virtual assistants like Siri and Alexa use female voices, reinforcing gender bias.
Recognition Failures: Computer vision struggles with recognizing people of different skin colors
accurately.
Module 4: Current Trends in Artificial Intelligence
8.1.4 AI and Bias
Sources of Bias in AI
Data Bias: The data used to train AI systems is often biased.
Example: Voice assistants may struggle with accents (e.g., Indian accent) because they are
trained mostly on American English data.
Algorithmic Bias: AI algorithms can amplify existing biases in the data.
Example: An image search for "teacher" may show more women because the data used
to train the algorithm has more images of female teachers.
Human Bias: The people creating AI systems also unintentionally introduce bias.
Example: Developers may unknowingly design biased algorithms or train models using
skewed data that reflects their own biases.
Module 4: Current Trends in Artificial Intelligence
8.1.4 AI and Bias
Why Does Bias Happen in AI?
AI reflects the biases in the data provided, and because humans feed data into AI systems, the
bias is often rooted in human decisions and preferences.
Ethics and bias in AI are ultimately issues created by people, not machines.
Module 4: Current Trends in Artificial Intelligence
8.1.5 Towards Ethical and Trustworthy AI
Challenges with AI in Recruitment:
Issue: Some AI hiring platforms show biased results, like women being disproportionately rejected
based on years of biased data in male-dominated sectors.
Solution: Addressing these biases by regulating AI practices and continuously monitoring recruitment
algorithms.

Regulating Ethical AI:

European Commission (April 2021(:
Launched a legal framework to ensure AI safety, protecting fundamental rights.
Introduced a risk-based approach to set requirements and ban AI systems that could harm people’s
rights, such as systems that manipulate human behavior.
Australia ( June 2021(:
Launched a similar AI ethics framework to guide businesses and governments on ethical AI usage.
Module 4: Current Trends in Artificial Intelligence
8.1.5 Towards Ethical and Trustworthy AI
Company and Organizational Engagement:
Building Trust: Organizations are implementing:
Cultural programs to raise awareness.
Risk assessments to identify potential AI issues.
Third-party audits to ensure compliance with ethical standards.
Framework: AI systems should be monitored using a prevention, detection, and response framework (like anti-
corruption measures).

Role of Rights and Activist Groups:

Human-Centric AI: AI systems must prioritize human rights and be developed with social responsibility.
Challenge: Civil rights groups should challenge AI practices and amplify the voices of those most affected by
technology.
Questions to Ask:
Is the technology necessary?
Is there an alternative?
Does the benefit outweigh the harm?
Module 4: Current Trends in Artificial Intelligence
8.1.5 Towards Ethical and Trustworthy AI
Ensuring Data Privacy:
Data Protection: Essential for ethical AI.
Encryption: Ensures data is secure.
Secure Computation: Prevents AI developers from accessing private data.
Dataset Boundaries: AI should not learn beyond its provided dataset.
Prevent Reverse Engineering: Ensures data used in AI models is inaccessible to users.

Diversify Your Team to Prevent Bias:

Diverse Teams: Forming teams with varied backgrounds (e.g., data scientists, business leaders, lawyers, sociologists,
and ethicists) can help in identifying and preventing bias in AI systems.
Fairness Checks: Regular analysis of data and algorithms ensures that biases are spotted early.
Bias Mitigation Tools: Tools like Bias Analyzer help automate the detection of bias and analyze the effectiveness of
mitigation actions.
Module 4: Current Trends in Artificial Intelligence
8.1.6 Why is Ethical AI Important?
Critical Areas of Application:
AI is used in medicine, law enforcement, recruitment, data privacy, military defense, and self-
driving vehicles.
These systems must produce accurate, transparent, and understandable results, aligned with
societal ethical standards.

Risks of Unethical AI:

Biased or incorrect outputs can:
Erode privacy (e.g., misusing personal data).
Lead to decisions that are impossible to understand.
Cause harm without clear accountability for damages.
Module 4: Current Trends in Artificial Intelligence
8.1.6 Why is Ethical AI Important?
Importance of Fairness & Accuracy:
AI must be unbiased and accurate
Without ethical design, AI systems can be misused, leading to disastrous consequences for
humanity.

Examples of Unethical AI Outputs:

Medical Bias: An algorithm used in US hospitals to predict patients needing extra care favored
white patients over black patients due to biased data regarding medical billing and facilities.
Hiring Bias: Amazon's AI hiring algorithm discriminated against women due to a data bias from
previous male-dominated applications.
Module 4: Current Trends in Artificial Intelligence
8.1.6 Why is Ethical AI Important?
Efforts to Ensure Ethical AI:
IEEE Global Initiative and other committees focus on making AI systems more ethical.
Ethically Aligned Design (2016(: A publication providing guidelines for ethical AI solutions.

Guiding Principles for Trustworthy AI (European Commission High-Level Expert Group(:

Support Human Autonomy: AI should enhance human decision-making.
Robust & Safe: AI must be technically sound with a fallback plan for failure.
Data Privacy: Protect user data and maintain data quality.
Transparency: AI should avoid biased data and be transparent about its models.
Fairness: Train AI systems with non-discriminatory, diverse data to avoid bias.
Benefiting Everyone: AI should promote societal and environmental wellbeing.
Accountability: AI must be responsible for accurate, unbiased outcomes.
Data Security: Ensure confidentiality, transparency, and safety in data management.
Accountability for Decisions: Ensure organizational responsibility for AI decisions.
Module 4: Current Trends in Artificial Intelligence
8.1.7 Impact of AI on Jobs
AI and Job Creation:
AI can displace 75 million jobs, but will create 133 million new ones in the coming years.
AI brings a paradigm shift in industries, similar to the Industrial Revolution.
Predicted Job Losses by 2030:
30% of jobs could be replaced by AI and automation, impacting 400-800 million jobs globally.
375 million people might need to switch to new job categories.
AI's Role in Job Automation:
Jobs that involve predictable, repetitive tasks are most likely to be automated, such as waiters in
restaurants (e.g., robot waiters).
AI will not make all human workers obsolete, especially for tasks requiring emotion, ethics, and
complex decision-making.
Module 4: Current Trends in Artificial Intelligence
8.1.7 Impact of AI on Jobs
Jobs That Will Grow with AI:
Creative Jobs: AI will enhance professions like artists, doctors, and scientists, making their work
more efficient.
Management Jobs: AI cannot replace human managers as managing people involves
understanding emotions, communication, and ethics.
Tech Jobs: Roles like programmers, data scientists, and big data engineers will be in high
demand as they work on creating and maintaining AI systems.

AI’s Dual Impact on Jobs:

AI will create new opportunities by removing routine, low-value jobs, replacing them with jobs
that require higher skills (programming, robotics, engineering).
Flexible and premium jobs will emerge, moving away from traditional blue-collar and white-
collar work.
Module 4: Current Trends in Artificial Intelligence
8.1.7 Impact of AI on Jobs
Emerging Job Titles by 2030:
New roles like Chief Bias Officer, Data Detective, AI Tutor, AI Business Development Manager
will emerge.

AI’s Impact on Sectors:

Medical: AI will help in predictive healthcare models, diagnosis, and reducing disease.
Automotive: Autonomous vehicles and manufacturing automation are powered by AI.
Cybersecurity: AI helps identify and predict cyber threats and fraud.
E-commerce: AI powers recommendation systems, chatbots, and inventory management.
HR: AI tools can filter resumes, saving time for HR professionals.
Legal: AI aids in document handling, classification, and knowledge extraction.
Module 4: Current Trends in Artificial Intelligence
8.2 AI as a Service (AIaaS(
8.2.1 Factors Triggering Growth of AIaaS
Cloud Platforms: Availability of affordable cloud platforms for enterprise data management.
Improved Data Storage: Cost-effective and reliable data storage technologies.
Data Generation: Streaming devices and IoT generate massive data that can be analyzed for
competitive advantage.
Automation Tools: Semi or fully automated data management and analytics tools simplify
business analysis.
Module 4: Current Trends in Artificial Intelligence
8.2 AI as a Service (AIaaS)
8.2.2 The Growth of AIaaS
Accessibility: Cloud services make AI available to more companies, reducing the need for massive
investment in talent and resources.
Key Benefits:
Focus on Core Business: Companies can focus on their core business instead of learning AI.
Cost Reduction: AIaaS lowers operational costs and avoids expensive infrastructure.
Minimized Investment Risk: No huge upfront costs or resource commitments.
Quick Deployment: AIaaS offers fast solutions with pre-built algorithms.
Data Insights: Gain competitive edge by analyzing trends and improving data usage.
Scalability: Companies can start small and scale AI projects as needed.
Module 4: Current Trends in Artificial Intelligence
8.2 AI as a Service (AIaaS(
8.2.3 Challenges of AIaaS
1. Security: Sharing data with third-party vendors can raise security concerns.
2. Reliance on Third Parties: Dependence on vendors for services can lead to delays.
3. Reduced Transparency: Lack of insight into the internal workings of AI systems.
4. Data Governance: Industry-specific data policies may restrict the use of third-party AIaaS.
Module 4: Current Trends in Artificial Intelligence
8.2 AI as a Service (AIaaS(
8.2.4 Vendors of AIaaS
Major AIaaS Providers:
Amazon Web Services (AWS(: Offers machine learning services like bots, APIs, and
frameworks.
Microsoft Azure: Provides AI and ML solutions, including cognitive services and bot services.
Google Cloud Platform (GCP(: Helps data scientists and developers create and deploy
machine learning projects.
Other AIaaS Providers:
MonkeyLearn: Simplifies text analysis using no-code tools.
IBM Watson: Offers a suite of AI tools for building virtual assistants and text analysis.
Microsoft Azure: Provides cognitive services for computer vision and text extraction.
Google Cloud ML: Facilitates data scientists in creating and deploying ML models.
Module 4: Current Trends in Artificial Intelligence
8.2 AI as a Service (AIaaS(
8.2.5 Market Growth
SaaS Market: Valued at $133 billion.
AI Platform-as-a-Service (AIPaaS(: Expected to grow from $11 billion in 2023 to $88.5 billion by
2025.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.1 Artificially Intelligent Robot
AI vs Robotics: AI and robotics are often confused, but they are not
synonymous. AI involves programming machines to simulate human-like
decision-making, whereas robotics involves designing and developing
machines that perform physical tasks.
Artificially Intelligent Robots (AIRs(: These robots bridge the gap
between AI and robotics. AIRs use AI algorithms to perform complex
tasks that require decision-making, such as:
Warehouse robots: Use path-finding algorithms for navigation.
Drones: Use autonomous navigation to return home when their
battery is low.
Self-driving cars: Detect and avoid hazards using AI.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.3 Types of Robots
1. Pre-Programmed Robots:
Perform simple, repetitive tasks in controlled environments.
Example: Robots on automotive assembly lines.
2. Humanoid Robots:
Mimic human behavior and actions (e.g., walking, carrying objects).
Example: Sophia by Hanson Robotics.
3. Autonomous Robots:
Operate independently without human intervention.
Examples: Roomba vacuum, autonomous drones, medical assistant bots.
4. Teleoperated Robots:
Semi-autonomous robots controlled remotely by humans.
Example: Drones used for landmine detection or fixing underwater pipe leaks.
5. Augmenting Robots (VR Robots(:
Enhance human abilities, like robotic prosthetics or exoskeletons.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.4 Types of Robots Based on Degree of Human
Control Independent Robots:
Operate autonomously to replace humans in dangerous or mundane tasks.
Example: Bomb disposal robots, deep-sea exploration robots.
Dependent Robots:
Require human interaction or guidance.
Example: Prosthetic limbs controlled by human signals.
Chatbots:
Software robots that simulate conversation and are commonly used in customer
service.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.5 Components of a Robot
1. Control System:
Acts as the robot’s brain, directing its tasks through computations.
2. Sensors:
Allow robots to interact with their environment by detecting changes and events.
Examples: Cameras (vision), microphones (sound), photoresistors (light).
3. Actuators:
Motive components that enable robot movement based on signals from the control system.
Examples: Electric motors for rotational movement.
4. Power Supply:
Powers robots using batteries or external sources (e.g., solar, hydraulic, or pneumatic power).
5. End Effectors:
Physical components (like robotic hands, grippers) that complete tasks.
Example: Gripping claws used in factories for handling materials.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.6 AI Technology Used in Robotics
Computer Vision: Extracts useful Transfer Learning: Utilizes pre-trained
information from images and videos for models for related tasks, reducing the training
robots to act on. time and cost.
Natural Language Processing (NLP(: Allows Reinforcement Learning: Robots learn
robots to understand and respond to human optimal behaviors through trial and error
language. based on feedback.
Edge Computing: Helps integrate robots with Affective Computing: Adds emotional
better data management and processing intelligence to robots to simulate human-like
capabilities. emotions.
Complex Event Processing (CEP(: Processes Mixed Reality: Combines physical and virtual
multiple real-time events, crucial for robotic worlds for interactive robotic programming
applications like healthcare or security. and demonstrations.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
8.3.7 Planning and Navigation
Cognition: Refers to a robot's ability to make purposeful decisions to achieve its goals. In
mobile robots, this means moving towards a goal efficiently and reliably.
Path Planning: Identifies the best route from the start to the goal, avoiding obstacles.

[Link] Competencies for Planning

Strategic Problem Solving: Involves planning the best trajectory to reach a goal.
Tactical Problem Solving: Involves avoiding obstacles and reacting to real-time data.

[Link] Key Terms in Trajectory Planning

Trajectory Planning: Moving from point A to point B while considering time, velocity, and
obstacles.
Configuration Space: The set of all possible positions a robot can occupy.
Free Space: The area in configuration space that the robot can navigate without collisions.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
[Link] Problem Constraints in Trajectory Planning
Holonomicity: Describes whether the robot has the ability to move freely in all directions.
Dynamic Environments: In real-world situations, obstacles may move, making trajectory
planning more challenging.

[Link] Planning Algorithms

1. Artificial Potential Field: Directs the robot to the goal while avoiding obstacles using an
attractive force for the goal and repulsive forces for obstacles.
2. Sampling-based Planning: Selects sample configurations and uses search algorithms to find a
path to the goal.
3. Grid-based Planning: Divides the environment into a grid, checking for collision-free paths to
the goal.
4. Reward-based Planning: Aims to maximize future rewards by selecting the best actions, often
used in reinforcement learning.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
[Link] Visibility Graph
Visibility Graph: Connects visible vertices in the environment to form a path, finding the shortest route
while avoiding obstacles.

[Link] Voronoi Diagram

Voronoi Diagram: Maximizes the distance between the robot and obstacles, helping in safe path
planning.

[Link] Cell Decomposition Path Planning

1. Exact Cell Decomposition: Divides the environment into cells and checks for free space.
2. Approximate Cell Decomposition: Uses fixed grid sizes for simpler and faster path planning, though it
may miss narrow passageways.
[Link] Potential Field Path Planning
Potential Field: Uses forces to guide the robot to its goal, similar to gravity, with attractive forces for the
goal and repulsive forces for obstacles.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
Topic Description
AI: Programming machines to simulate human decision-making. Robotics:
AI vs Robotics
Designing machines to perform physical tasks. AIRs: Combination of both.
Artificially Robots controlled by AI algorithms to perform complex tasks. Examples:
Intelligent Robots warehouse robots, drones, self-driving cars.

Characteristics of 1. Electrical components for control and power. 2. Control systems with
Robots programmed instructions for tasks.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
1. Pre-programmed Robots: Simple tasks (e.g., industrial robots). 2.
Humanoid Robots: Mimic human behavior (e.g., Sophia). 3. Autonomous
Types of Robots Robots: Operate independently (e.g., Roomba). 4. Teleoperated Robots:
Controlled remotely (e.g., drones for landmine detection). 5. Augmenting
Robots: Enhance human capabilities (e.g., robotic prosthetics).
1. Independent Robots: Fully autonomous (e.g., bomb disposal robots). 2.
Types of Robots
Dependent Robots: Require human guidance (e.g., prosthetics controlled by
(Human Control)
signals). 3. Chatbots: Software robots for conversation.
1. Control System: Brain of the robot. 2. Sensors: Detect environmental
Components of a changes (e.g., cameras, microphones). 3. Actuators: Enable movement (e.g.,
Robot motors). 4. Power Supply: Provides energy (e.g., batteries). 5. End Effectors:
Components to complete tasks (e.g., grippers, hands).
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
1. Computer Vision: Extracts information from visual inputs. 2. NLP: Allows
voice commands. 3. Edge Computing: Enhances data processing. 4. Complex
AI Technology in
Event Processing (CEP): Handles multiple real-time events. 5. Transfer
Robotics
Learning: Uses previous knowledge for related tasks. 6. Reinforcement
Learning: Learns from actions and feedback. 7. Affective Computing:
Path planning for efficient robot movement. Cognition: Decision-making
Planning and
capability to reach goals. Planning: Strategically navigate obstacles, Reacting:
Navigation
Adjusts based on sensor feedback.

Competencies for
1. Strategic: Make decisions to reach goals. 2. Tactical: Avoid obstacles.
Planning
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
Topic Description

1. Trajectory Planning: Moving with velocity, time, and obstacles in

Key Terms in Trajectory mind. 2. Configuration Space: Set of all possible positions a robot can
Planning occupy. 3. Free Space: Areas without obstacles. 4. Target Space:
Desired area robot aims to navigate.
Holonomicity: Relation between controllable and total degrees of
Problems in Trajectory
freedom. Dynamic Environments: Moving obstacles make planning
Planning
more challenging.
1. Artificial Potential Field: Robot moves based on attractive and
repulsive forces. 2. Sampling-based Planning: Uses milestones to find
Planning Algorithms
paths. 3. Grid-based Planning: Uses grid to plan paths. 4. Reward-
based Planning: Uses rewards for optimal decisions.
Module 4: Current Trends in Artificial Intelligence
8.3 Robotics
Topic Description

Connects visible points in the environment to form a path, finding

Visibility Graph
the shortest route.

Voronoi Diagram Maximizes distance between robot and obstacles for safe navigation.

1. Exact Cell Decomposition: Divides space into cells for path

Cell Decomposition
planning. 2. Approximate Cell Decomposition: Uses fixed grids for
Path Planning
faster planning.

Potential Field Path Uses artificial forces (attractive for goals, repulsive for obstacles) to
Planning guide the robot towards its goal.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.1 Collaborative Systems
Machine-Human Collaboration:
AI and humans work together to complement each other's strengths.
Examples:
Foldit: AI and humans collaborate to fold proteins.
AI in Chess: Amateurs and AI beat grandmasters.
AI in Business: AI elected to company boards.
Military: Drones and humans working together.
Exoskeletons: Power suits assist soldiers in the field.
Challenges:
Over-reliance on AI could lead to job loss and skill degradation.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.2 Machines Assisting Humans
AI Helps Humans Enhance Abilities:
AI boosts creativity, decision-making, and interactions.
Examples:
SEB Bank’s AI Assistant: Helps customers with tasks and understands tone.
AI improves efficiency in various sectors like healthcare, customer service, and homes.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.3 Algorithmic Game Theory
Game Theory with AI:
AI helps machines play games ethically using predefined rules.
Types of Games:
Cooperative vs. Non-cooperative: Alliances or no alliances.
Symmetric vs. Asymmetric: Same vs. different goals.
Perfect vs. Imperfect Info: Full visibility vs. hidden moves.
Zero-sum vs. Non-zero sum: One loses, one wins vs. all gain.
Key Concepts:
Nash Equilibrium: No player can improve their situation by changing strategies.
Inverse Game Theory: Designing games based on players’ strategies.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.4 Multi-Agent Reinforcement Learning (MARL(
Multiple Agents Learning:
Multiple AI agents interact and learn together.
Example:
Self-driving cars: Cars learning to manage traffic flow.
Challenges:
Increased complexity with more agents interacting.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.5 Neuromorphic Computing
Brain-like AI Systems:
Neuromorphic computing mimics the brain’s structure for efficient processing.
Features:
Parallel processing, low power consumption, flexibility, fault-tolerance.
Applications:
Real-time processing, energy-efficient systems, adaptive AI.
Difference:
Uses spiking neural networks (SNNs) for processing.
Examples:
Intel’s Loihi and IBM’s TrueNorth chips.
Module 4: Current Trends in Artificial Intelligence
8.4 Recent Trends in AI
8.4.6 Challenges in Neuromorphic Computing
Difficult Transition:
Moving from Von Neumann architecture to neuromorphic computing is challenging.
Needs:
New programming languages, sensors, and storage technologies.
Programming:
Requires new development for neuromorphic systems.

Impact of AI Trends
Mainstream AI:
AI is becoming standard across industries.
New opportunities with challenges in ethical implementation.
 Module 4: Current Trends in Artificial Intelligence
Chapter 8 Current Trends in Artificial Intelligence
8.1 AI and Ethical Concerns 8.2 AI as a Service (AIaaS( 8.3 Robotics 8.4 Recent Trends in AI
8.1.1 Ethical Use of Artificial 8.2.1 Factors Triggering 8.3.1 Artificially Intelligent Robot 8.4.1 Collaborative Systems
Intelligence Growth of AIaaS 8.3.2 Characteristics of Robots 8.4.2 Machines Assisting Humans
8.1.2 Is AI Dangerous? Will 8.2.2 The Growth of AIaaS 8.3.3 Types of Robots 8.4.3 Algorithmic Game Theory and
Robots Take Over the World? 8.2.3 Challenges of AIaaS 8.3.4 Types of Robots Based on Computational Social Choice
8.1.3 Ethics in AI 8.2.4 Vendors of AIaaS Degree of Human Control 8.4.4 Multi-Agent Reinforcement
8.1.4 AI and Bias 8.3.5 Components of a Robot Learning (MARL)
8.1.5 Towards Ethical and 8.3.6 AI Technology Used in 8.4.5 Neuromorphic Computing
Trustworthy AI Robotics
8.1.6 Why is Ethical AI 8.3.7 Planning and Navigation
Important?
8.1.7 Impact of AI on Jobs
Module 4: Current Trends in Artificial Intelligence
Chapter 9 Where AI Is Heading Today?
9.1 Expert System 9.2 Internet of Things 9.3 Artificial Intelligence of Things
9.1.1 Popular Examples of the Expert System 9.2.1 Examples of Applications of IoT (AIoT(
9.1.2 Characteristics of an Expert System 9.2.2 IoT Products 9.3.1 How Does AIoT Work?
9.1.3 Components of an Expert System 9.2.3 Challenges 9.3.2 Where Does AI Unlock IoT?
9.1.4 Participants in the Development of 9.2.4 Sensors 9.3.3 Applications and Examples of AIoT
Expert System 9.3.4 Benefits and Challenges of AIoT
9.1.5 Capabilities of the Expert System 9.3.5 Future of AIoT
9.1.6 Advantages of Expert Systems
9.1.7 Limitations of Expert Systems
9.1.8 Applications of Expert Systems
9.1.9 Expert System Technology
9.1.10 Development of Expert Systems
Module 4: Current Trends in Artificial Intelligence
9.1 Expert System
A computer program designed to solve complex problems with human-like expertise.
Used to assist, not replace, human decision-making in specific domains (e.g., medicine, science).
Key Components:
a. User Interface: Allows interaction with the system.
b. Knowledge Base (KB): Stores expert knowledge for decision-making.
c. 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.
Module 4: Current Trends in Artificial Intelligence
9.1.1 Popular Examples of Expert Systems
DENDRAL:
A chemical analysis system.
Used in organic chemistry to identify unknown molecules by analyzing mass spectra.
MYCIN:
Designed to diagnose blood clotting diseases and bacterial infections.
Recommends antibiotics and drugs based on the patient's condition.
PXDES:
Determines the type and level of lung cancer using images of the upper body.
CaDeT:
Diagnostic support system to detect cancer at early stages.
R1/XCON:
Customizes computer systems based on user requirements.
DXplain:
Clinical support system that suggests diseases based on doctor’s findings.
Module 4: Current Trends in Artificial Intelligence
9.1.2 Characteristics of an Expert System
High Performance:
Solves complex domain problems with high efficiency and accuracy.
Dependent on a well-updated knowledge base.
Understandable:
Results are provided in a human-understandable format.
Takes input and gives output in natural language.
Reliable:
Generates efficient and accurate output consistently.
Highly Responsive:
Provides fast results, faster than human experts.
No Memory Limitations:
Can store and process huge amounts of data.
Module 4: Current Trends in Artificial Intelligence
9.1.2 Characteristics of an Expert System
Expertise in a Domain:
Integrates knowledge from multiple human experts in specific areas.
Not Affected by Emotions:
Consistent performance without the influence of fatigue, stress, or emotions.
Not Biased:
Checks all facts before making decisions, ensuring impartial results.
Reduce Cost:
Much cheaper than hiring human experts for repeated consultations.
Non-perishable:
Can be used indefinitely once developed, unlike human experts.
Intelligent:
Uses knowledge and inference rules to deduce new facts and solve problems.
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
An expert system mainly consists of three
components—
1. User interface,
2. Inference engine and
3. Knowledge base
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
1. User Interface
Role: Allows users to interact with the system.
Function: Accepts user queries in natural language and provides understandable results.
Key Features:
Helps users accomplish tasks quickly.
Customizable for user requirements.
Efficiently handles user input.
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
2. Inference Engine (Rules Engine(
Role: The "brain" of the expert system.
Function: Processes information from the knowledge base to draw conclusions.
Types:
Deterministic: Draws conclusions with certainty.
Probabilistic: Deals with uncertainty and probability.
Strategies:
Forward Chaining: Starts with known facts to predict what can happen next (e.g.,
predicting stock prices).
Backward Chaining: Starts with the goal to work backward and understand the cause (e.g.,
diagnosing diseases).
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
3. Knowledge Base
Role: Stores domain-specific knowledge.
Types of Knowledge:
Factual Knowledge: Based on facts accepted in the domain.
Heuristic Knowledge: Based on experience, probability, and practices.
Representation: Knowledge is typically represented using If-Else Rules.
Case Study: MYCIN (Expert System(
Step 1: Knowledge base about bacterial infections is created.
Step 2: The system is updated with new information.
Step 3: The doctor inputs details of a patient (e.g., symptoms, medical history).
Step 4: The system applies If-Then rules using the inference engine.
Step 5: Output is provided through the user interface, guiding diagnosis and treatment.
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
Inference Engine (Rules Engine(
The Inference Engine is the core processing unit of an expert system, often referred to as its "brain."
It processes knowledge from the knowledge base and draws conclusions based on inference rules.

Types of Inference Engines:

1. Deterministic Inference Engine:
How it works: Uses facts and rules to draw conclusions.
Output: Conclusions are assumed to be true based on the given data.
2. Probabilistic Inference Engine:
How it works: Deals with uncertainty and draws conclusions based on probability.
Output: Conclusions have some level of uncertainty attached.
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
Inference Engine (Rules Engine(
Functions of Inference Engine:
Efficient Processing: Applies rules and procedures to generate correct solutions.
Conflict Resolution: If multiple rules apply, the engine selects the best rule based on the
current context.
Explanation and Debugging: Can explain how conclusions are reached, ensuring
transparency.
Module 4: Current Trends in Artificial Intelligence
9.1.3 Components of an Expert System
Inference Engine (Rules Engine(
Strategies for Knowledge Acquisition:
Forward Chaining:
How it works: Starts with known facts and
rules to predict what can happen next.
Example: Predicting stock prices after
interest rate changes.
Backward Chaining:
How it works: Starts from the goal and
works backward to prove the facts that lead
to it.
Example: Diagnosing blood cancer by
tracing the cause from the symptoms.
Module 4: Current Trends in Artificial Intelligence
9.1.4 Participants in the Development of Expert System

Expert:
Role: Provide the domain-specific knowledge that feeds into the expert system's knowledge base.
Importance: The accuracy and performance of the expert system rely on the quality and depth of knowledge
provided by the experts.
Knowledge Engineer:
Role: Responsible for building the expert system by acquiring, organizing, and structuring knowledge.
Importance: The knowledge engineer ensures the knowledge is represented in a format that the expert system
can use effectively to solve problems.
End User:
Role: The person who interacts with the expert system to get solutions or advice for specific queries.
Importance: They may not have deep knowledge in the domain but rely on the expert system to address complex
issues or tasks.
Module 4: Current Trends in Artificial Intelligence
9.1.5 Capabilities of the Expert System
Advise: Provides advice for any query within Predict Results: Predicts outcomes based on
its domain. historical data.
Decision-Making: Makes complex decisions Assist in Decision-Making: Assists humans
in fields like finance, medicine, etc. in making informed decisions.
Demonstrate Devices: Demonstrates Diagnose: Used in medical fields to
features, specifications, and usage of new diagnose diseases.
products. Justify Conclusions: Explains the reasoning
Problem-Solving: Solves complex problems behind the conclusions drawn.
within its domain. Suggest Alternatives: Suggests alternative
Explain a Problem: Offers detailed solutions to problems.
descriptions of input problems.
Interpret Input: Interprets user input for
further processing.
Module 4: Current Trends in Artificial Intelligence
9.1.5 Capabilities of the Expert System

Limitations of the Expert System

Cannot Substitute Human Decision-Makers: It does not replace human judgment.

Lacks Human Capabilities: Cannot replicate human emotions or intuitive thinking.
Inaccurate Output with Inadequate Knowledge Base: Performance depends on the
quality of the knowledge base.
Cannot Refine Knowledge Base: It cannot autonomously improve or update its
knowledge.
Cannot Use Emotions in Decisions: Decisions are purely logical, without emotional
influence.
Module 4: Current Trends in Artificial Intelligence
9.1.6 Advantages of Expert Systems
Reproducible Results: Results are consistent Affordable: Production costs are reasonable,
and can be reproduced each time. making them cost-effective.
Safe in Risky Environments: Can be used in High Speed: Operate at high speed,
hazardous places where human presence is reducing human effort and time.
risky. Explainable Results: Capable of explaining
Accuracy: Provide accurate results, especially how conclusions or solutions were reached.
with regularly updated knowledge bases. Improved Decision Quality: Enhance the
Consistent and Fast Performance: They quality of decisions made.
perform consistently without being affected Efficient Use of Expert Knowledge:
by emotions or fatigue. Efficiently gather and use scarce expertise
Easily Available: Mass-produced software for better solutions.
makes expert systems widely available.
Module 4: Current Trends in Artificial Intelligence
9.1.7 Limitations of Expert Systems
Wrong Results: Incorrect or outdated knowledge leads to Training Intensive: Requires extensive training to
wrong results. function well.
Different Solutions: Human experts may give different No Emotions: Cannot understand or use emotions in
answers from the system. decisions.
Lack of Creativity: Can't handle unusual or creative No Common Sense: Lacks human-like reasoning.
scenarios. Domain-Specific: Only works for a specific area.
High Cost: Expensive to develop and maintain. Struggles with Unusual Situations: Can't make decisions
Hard to Gather Knowledge: Collecting the right in rare cases.
knowledge is difficult. Garbage In, Garbage Out (GIGO(: Poor data leads to
Depend on Experts: Success relies on expert knowledge poor results.
input. High Maintenance Cost: Regular updates and
No Self-Learning: Can't learn from experience, needs maintenance are expensive.
manual updates.
Needs Plenty of Knowledge: Can't work with limited
information.
Module 4: Current Trends in Artificial Intelligence
9.1.8 Applications of Expert Systems
Design & Manufacturing: VLSI systems, camera Software Development: Tracking project
lenses, automobiles. progress.
Knowledge Publishing: Tax advisors and other Monitoring Systems: Detect leaks in pipelines,
advisory systems. continuous data comparison.
Finance: Detect fraud, suspicious activities, stock Process Control: Control physical processes
trading, loan approvals. through monitoring.
Medical Diagnosis: Used extensively for Automobile & Electronics: Fault detection in
diagnosing diseases and treatment. vehicles and computers.
Planning & Scheduling: Project goals, airline Complex System Diagnosis: Electronic,
scheduling, cargo, manufacturing. electromechanical, diesel-electric systems.
Module 4: Current Trends in Artificial Intelligence
9.1.8 Applications of Expert Systems
Design & Manufacturing: VLSI systems, camera Software Development: Tracking project
lenses, automobiles. progress.
Knowledge Publishing: Tax advisors and other Monitoring Systems: Detect leaks in pipelines,
advisory systems. continuous data comparison.
Finance: Detect fraud, suspicious activities, stock Process Control: Control physical processes
trading, loan approvals. through monitoring.
Medical Diagnosis: Used extensively for Automobile & Electronics: Fault detection in
diagnosing diseases and treatment. vehicles and computers.
Planning & Scheduling: Project goals, airline Complex System Diagnosis: Electronic,
scheduling, cargo, manufacturing. electromechanical, diesel-electric systems.
Module 4: Current Trends in Artificial Intelligence
9.1.9 Expert System Technology
Expert System Development Environment:
Hardware: Workstations, minicomputers, mainframes.
Programming Languages: LISP, PROLOG (high-level symbolic programming).
Databases: Large databases for storing information.
Tools:
Powerful Editors: Help developers efficiently create systems.
Debugging: Multi-window debugging for easy identification of errors.
Rapid Prototyping: Fast development and testing.
Predefined Models: In-built knowledge representation and inference design.
Shells:
Definition: Expert systems without a knowledge base, providing essential modules.
Modules: Knowledge acquisition, inference engine, user interface, and explanation.
Popular Shells:
JESS: Java Expert System Shell, used with Java API.
Vidwan: Enables encoding knowledge in IF-THEN rules.
Module 4: Current Trends in Artificial Intelligence
9.1.10 Development of Expert Systems
Identify Problem Domain: Test and Refine the Prototype:
Identify the problem to solve. Test the prototype with sample cases.
Ensure it is solvable by an expert system. Work with end-users to find errors and discrepancies.
Collaborate with human experts and assess cost- Refine the system based on test results.
effectiveness.
Develop and Complete the System:
Design the System: Test the system with end-users and databases.
Choose hardware, tools, and shells. Integrate the system with other information systems.
Plan system integration with other systems and Document the system and train users.
databases.
Design a framework for representing knowledge. Maintain the System:
Regularly review and update the knowledge base.
Develop the Prototype: Integrate new requirements and interfaces as systems
Acquire domain knowledge from experts. evolve.
Represent knowledge in If-THEN-ELSE rules.
Integrate heuristic knowledge for reasoning.
Module 4: Current Trends in Artificial Intelligence
9.1.10 Development of Expert Systems
Identify Problem Domain: Test and Refine the Prototype:
Identify the problem to solve. Test the prototype with sample cases.
Ensure it is solvable by an expert system. Work with end-users to find errors and discrepancies.
Collaborate with human experts and assess cost- Refine the system based on test results.
effectiveness.
Develop and Complete the System:
Design the System: Test the system with end-users and databases.
Choose hardware, tools, and shells. Integrate the system with other information systems.
Plan system integration with other systems and Document the system and train users.
databases.
Design a framework for representing knowledge. Maintain the System:
Regularly review and update the knowledge base.
Develop the Prototype: Integrate new requirements and interfaces as systems
Acquire domain knowledge from experts. evolve.
Represent knowledge in If-THEN-ELSE rules.
Integrate heuristic knowledge for reasoning.
Module 4: Current Trends in Artificial Intelligence
9.2 Internet of Things
IoT is a system of interrelated computing devices, machines, objects, animals, or people with unique identifiers
that can transfer data over a network without human-to-human or human-to-computer interaction.

Examples of IoT Devices:

People: Heart monitor implants.
Animals: Biochip transponders.
Automobiles: Sensors for monitoring tyre pressure.
Objects: Any device assigned an IP address to transfer data.

Applications of IoT:
Precision Agriculture: Monitoring crops, soil, and environment.
Building Management: Automating lighting, heating, and security systems.
Healthcare: Remote patient monitoring with wearable devices.
Energy: Smart grids and energy management systems.
Transportation: Traffic management, vehicle monitoring, etc.
Module 4: Current Trends in Artificial Intelligence
9.2.1 Examples of Applications of IoT
Smart Car Navigation:
Car suggests the best route, updates on traffic, and sends notifications if delayed.
Smart Home Automation:
Alarm rings, and coffee machine starts brewing automatically.
Wearable Fitness Trackers:
Smartwatch tracks activity and productivity, providing insights.
Self-Driving Cars:
Equipped with sensors for detecting objects and ensuring safe navigation.
Smart Football:
Tracks distance and speed, records stats for training.
Smart Refrigerator:
Notifies when groceries run out or when food is expired.
Smart Air Conditioner:
Controlled remotely via smartphone to set the desired temperature before arrival
Smart Cities:
IoT helps in reducing traffic congestion, managing resources efficiently, and reducing pollution and crime.
Module 4: Current Trends in Artificial Intelligence
9.2.2 IoT Products
Amazon Echo (Alexa(: Popular IoT Platforms:
Voice assistant for smart homes. Amazon Web Services (AWS)
Can perform various tasks like playing Microsoft Azure
music, providing weather reports, or IBM Watson
ordering an Uber. Cisco IoT Cloud Connect
Fitbit One (Wearables(: Salesforce IoT Cloud
Tracks physical activities (steps, calories, Oracle Integrated Cloud
sleep). GE Predix
Syncs with smartphones and computers to AT&T - Connected Car:
monitor progress. Added 1.3 million connected cars to its
Barcelona - Smart City: network in 2016.
Implemented IoT solutions for smart Brings the total to 9.5 million connected
parking and environmental improvements. cars.
One of the leading smart cities globally.
Module 4: Current Trends in Artificial Intelligence
9.2.3 Challenges in IoT
Security:
Privacy and data sharing concerns.
Hackers target connected devices (e.g., cars, homes).
Lack of encryption in communication can cause security vulnerabilities.
Scalability:
Huge volumes of data generated need big data analytics and cloud storage.
Devices must be scalable to process and store ever-increasing data.
Interoperability:
Lack of common standards for IoT devices.
Difficulty integrating with legacy systems.
Lack of Standardization:
Standardization issues hinder IoT device compatibility.
Prevents achieving truly connected systems.
Lack of Government Support:
Regulatory bodies (e.g., FDA) lack IoT-specific safety and security regulations.
Limited governmental oversight.
Module 4: Current Trends in Artificial Intelligence
9.2.3 Challenges in IoT
IoT Malware and Ransomware:
IoT devices vulnerable to malware and ransomware attacks.
Hackers can take control of devices and demand ransom.
Connectivity:
Poor internet connectivity affects IoT device performance.
IoT devices require stable connections for effective monitoring and control.
Privacy Policies:
Lack of clear regulations on what data is private or public.
Example: IoT toys collecting children’s data without parental consent.
Bandwidth Strain:
Growing number of IoT devices creates strain on network bandwidth.
Increased data flow demands higher bandwidth and powerful servers.
IoT Skills Gap:
Shortage of trained professionals in IoT network management.
Difficulty in training new staff due to rapid IoT development.
Module 4: Current Trends in Artificial Intelligence
9.2.4 Sensors
Sensor Type Description Example

Oxygen Sensor Detects the gasoline/oxygen ratio Car emission control system

Motion Sensor Detects interruption in energy flow Home security lights, automatic doors
Vision and Imaging
Detects objects or colors in view Cameras for object detection
Sensor
Temperature Sensor Measures thermal parameters Industrial processes for gases, liquids, and solids

Radiation Sensor Detects alpha, beta, or gamma Surveys and sample counting

Proximity Sensor Detects presence of objects without contact Manufacturing operations

Pressure Sensor Detects force per unit area in gases/liquids Control and display devices

Position Sensor Senses position of objects like valves or doors Control systems for machinery

Photoelectric Sensor Detects objects and can sense color Manufacturing and material handling
Module 4: Current Trends in Artificial Intelligence
9.2.4 Sensors
Sensor Type Description Example

Particle Sensor Detects airborne particulates Bin and baghouse monitoring

Motion Sensor Senses movement or stoppage Conveyor detection, stalling bearings

Metal Detector Senses the presence of metal Security checks

Level Sensor Determines height of substances in tanks Control and display systems

Leak Sensor Detects unwanted discharge of liquids/gases Monitoring gas leaks

Humidity Sensor Measures water content in the air Control systems for temperature regulation

Gas and Chemical Detects gases/chemicals Environmental monitoring

Force Sensor Measures forces like weight or torque Control systems for load measurement
Flow Sensor Detects movement of gases/liquids/solids Industrial processing systems
Flaw Sensor Detects inconsistencies on surfaces Manufacturing processes like welding
Module 4: Current Trends in Artificial Intelligence
9.2.4 Sensors
Sensor Type Description Example

Flame Detector Senses fire presence and quality Combustion control systems
Detects electrical parameters like current and
Electrical Sensor Monitoring electrical systems
voltage
Contact Sensor Detects physical touch or contact Alarm systems for doors/windows

Non-contact Sensor Does not require physical touch Radar guns for speed detection

Speed Sensor Measures speed of an object/vehicle Vehicle monitoring

Measures distance using high-frequency sound

Ultrasonic Sensor Object distance measurement in robotics
waves
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
AIoT combines Artificial Intelligence (AI) with the Internet of Things (IoT) to enhance efficiency,
improve decision-making, and enable better data management.
AI simulates human intelligence, while IoT connects devices to transfer data over networks without
human intervention.
AIoT improves IoT by adding machine learning algorithms to enhance decision-making.

9.3.1 How Does AIoT Work?

AI is embedded in IoT devices and connected through IoT networks.
Devices collect data, which is analyzed using AI techniques to improve productivity and efficiency.
Edge AI reduces bandwidth needs and minimizes delays in data processing.
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
9.3.2 Where Does AI Unlock IoT?
AI enhances the Analysis step in IoT, which directly impacts efficiency.
AIoT helps:
Manage, analyze, and obtain meaningful insights from data.
Provide fast, accurate data analysis.
Balance localized and centralized intelligence.
Ensure data privacy and security.
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
9.3.3 Applications and Examples of AIoT
AIoT is widely used in various fields:
Smart Cities: Enhances operational efficiency, improves traffic management, and optimizes
resource use.
Smart Retail: Uses smart cameras to track customers and optimize store operations.
Smart Homes: Devices analyze user habits to provide customized support.
Manufacturing: Monitors equipment and detects malfunctions.
Autonomous Vehicles: Uses real-time data to make driving decisions.
Healthcare: Devices monitor patient data, enabling remote care and accurate diagnoses.
Smart Thermostats: AI-powered devices adjust temperature based on user preferences and
schedules.
Fleet Management: Monitors vehicles, tracks fuel usage, and identifies driver behavior.
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
9.3.4 Benefits and Challenges of
AIoT Benefits:
Increases operational efficiency by analyzing and adjusting system operations in real-time.
Enables on-the-fly decision-making with real-time data analysis.
Reduces workload and saves time by automating data analysis.
Scalable systems that connect more devices as needed.
Better risk management by predicting potential problems in advance.
Improves products and services through real-time data analysis.
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
9.3.4 Benefits and Challenges of AIoT
Benefits:
Increases operational efficiency by analyzing and adjusting system operations in real-time.
Enables on-the-fly decision-making with real-time data analysis.
Reduces workload and saves time by automating data analysis.
Scalable systems that connect more devices as needed.
Better risk management by predicting potential problems in advance.
Improves products and services through real-time data analysis.
Challenges:
Security: Protection of sensitive data is a significant concern.
Scalability: Managing large volumes of data from multiple devices.
Interoperability: Lack of standardization among different IoT devices.
Privacy: Handling user data and ensuring privacy compliance.
Module 4: Current Trends in Artificial Intelligence
9.3 Artificial Intelligence of Things (AIoT(
9.3.5 Future of AIoT
Growth with 5G: 5G will boost AIoT by providing faster data transfer and lower latency.
Revolutionizing Industries: AIoT will improve supply chains, operational efficiency, and decision-
making.
Exponential Data Growth: With more devices connected, AI will help manage the massive
amount of data generated.
By 2025, the number of IoT devices is expected to rise to 42 billion, and AI will help process
and analyze the increasing data efficiently.
Module 4: Current Trends in Artificial Intelligence
Chapter 9 Where AI Is Heading Today?
9.1 Expert System 9.2 Internet of Things 9.3 Artificial Intelligence of Things
9.1.1 Popular Examples of the Expert System 9.2.1 Examples of Applications of IoT (AIoT(
9.1.2 Characteristics of an Expert System 9.2.2 IoT Products 9.3.1 How Does AIoT Work?
9.1.3 Components of an Expert System 9.2.3 Challenges 9.3.2 Where Does AI Unlock IoT?
9.1.4 Participants in the Development of 9.2.4 Sensors 9.3.3 Applications and Examples of AIoT
Expert System 9.3.4 Benefits and Challenges of AIoT
9.1.5 Capabilities of the Expert System 9.3.5 Future of AIoT
9.1.6 Advantages of Expert Systems
9.1.7 Limitations of Expert Systems
9.1.8 Applications of Expert Systems
9.1.9 Expert System Technology
9.1.10 Development of Expert Systems
 Thank You
Module 4 - Completed