PG Diploma in AI & Robotics Engineering – Tamil Nadu’s First 8-Month Program

Part 1: Foundations of Electronics & Circuitry

• Electrical Fundamentals: Basic circuit analysis, Ohm’s law, Kirchhoff’s laws (KCL/KVL), voltage dividers, power, and energy calculation.
• Passive & Active Components: Resistors, capacitors, PN-junction & Zener diodes, and transistors (BJT & MOSFETs as switches).
• Digital Electronics: Number systems (Binary/Hexadecimal), logic gates (AND, OR, NOT, NAND).
• Hands-on: Multimeter circuit testing, PCB layout concepts, and transistor switching circuits with LED indicators.

Part 2: Microcontroller Architecture & Programming

• Microcontroller Architecture: Microprocessor vs. microcontroller comparison, ARM/AVR architectures, registers, GPIOs, timers, and ADC/DAC concepts.
• Platform Setup (Arduino): Pin mapping, development board interfacing, and C/C++ syntax fundamentals using Arduino IDE.
• Digital & Analog I/O: Reading digital switches, debouncing logic, and Pulse Width Modulation (PWM) for motor speed and LED dimming.

Part 3: Sensor & Actuator Integration

• Serial Communication Protocols: Hardware wiring, timing, and protocol troubleshooting for UART, SPI, and I2C.
• Sensor Interfacing: Interfacing HC-SR04 ultrasonic, IR, PIR, DHT11 temperature/humidity, gas (MQ series), water level, and accelerometer/IMU sensors.
• Actuator Control: Driving DC motors via L298N/DRV8825 drivers, servo motor positioning, stepper motors, and relays.
• Hands-on: Obstacle detection distance monitor with live display readout and PWM motor speed control.

Part 1: Core Python & Object-Oriented Programming

• Introduction to Python: Environment setup, variables, data types, and control flow.
• Functions & Modules: Writing reusable code, lambda functions, and standard libraries.
• Object-Oriented Programming (OOP): Classes, objects, inheritance, and polymorphism.
• Hands-on: Mini Project.

Part 2: Mathematical Computing with NumPy

• Introduction to N-dimensional arrays and tensors.
• Indexing, slicing, and broadcasting.
• Linear algebra operations including dot products and matrix multiplication for robotic kinematics.
• Handling arrays for image and sensor data processing.

Part 3: Data Manipulation with Pandas

• Introduction to Series and DataFrames.
• Data cleaning, handling missing values, and filtering.
• Working with time-series data for analyzing continuous sensor feeds.
• Merging, joining, and aggregating datasets.

Part 4: Data Visualization with Matplotlib & Seaborn

• Creating line charts, scatter plots, and histograms.
• Visualizing trajectories and loss curves during model training.
• Plotting 3D data points for introductory spatial visualization.
• Hands-on: Exploratory Data Analysis (EDA).

Part 1: Machine Learning Foundations

• Introduction to AI: History, landscape, and AI’s role in modern robotics.
• Supervised Learning: Regression and Classification.
• Unsupervised Learning: Clustering algorithms and dimensionality reduction.

Part 2: Deep Learning Architectures

• Introduction to Deep Learning: Artificial neurons and activation functions.
• Core Architectures: Perceptron, Multi-Layer Perceptron (MLP), and Feedforward Neural Networks (FNN).
• Training Neural Networks: Backpropagation, loss functions, and optimization algorithms (Adam, SGD).
• Vision & Sequence Models: Convolutional Neural Networks (CNNs) for spatial data.
• Recurrent Neural Networks (RNNs & LSTMs) for sequential and time-series sensor data.

Part 3: Reinforcement Learning for Control

• Introduction to Reinforcement Learning: Markov Decision Processes (MDP), Agents, Environments, Rewards, and Policies.
• Value-Based Methods: Q-Learning and Deep Q-Networks (DQN).
• Policy & Imitation Learning: Proximal Policy Optimization (PPO) and Behavior Cloning.

Part 4: Computer Vision & Perception

• Introduction to Computer Vision: Image processing basics and OpenCV integration.
• Object Detection & Segmentation: Bounding boxes, semantic segmentation, and instance segmentation.
• YOLO (You Only Look Once): Architecture overview and pre-trained YOLO models for real-time perception.
• Customizing Models: Fine-tuning YOLO on custom robotics datasets.
• Modern Vision Architectures: Introduction to Vision Transformers (ViT) and attention mechanisms.

Part 1: ROS Architecture & Core Concepts

• Introduction to ROS Environment: Linux/Ubuntu terminal navigation, ROS workspace setup, package structure, and launch files.
• Communication Paradigms: ROS nodes, topics, publish/subscribe, custom message creation, services, and actions.
• CLI & GUI Developer Tools: Command-line inspection tools and RViz 3D visualization.

Part 2: Microcontroller Integration & Hardware Bridge

• Embedded ROS Bridge: micro-ROS and rosserial package architecture and configuration.
• Sensor Telemetry Pipeline: Publishing microcontroller sensor data including ultrasonic, IR, encoders, and IMU to ROS topics.
• Actuator Control Pipeline: Subscribing to ROS velocity command topics (cmd_vel) to drive physical DC/servo motors.

Part 3: Robot Modeling, Kinematics & Physics Simulation

• Robot Description Format: URDF and Xacro modeling, defining links, joints, visual, collision, and inertial properties.
• Coordinate Transformations: ROS TF2 package, dynamic and static parent-child coordinate transformations.
• Gazebo Physics Simulator: Creating 3D simulation worlds, spawning URDF models, and attaching differential-drive and sensor plugins.

Part 4: Robotic Manipulators & Motion Planning (MoveIt)

• Manipulator Kinematics: Degrees of Freedom (DoF), Forward Kinematics, and Inverse Kinematics.
• MoveIt Framework: MoveIt Setup Assistant configuration, path planning, trajectory execution, and obstacle avoidance (RRT, PRM).

Part 5: Autonomous Mobile Robots, SLAM & Navigation

• Occupancy Grid Mapping: Processing LIDAR and depth sensor streams into 2D maps.
• SLAM (Simultaneous Localization & Mapping): Map generation using Cartographer and Gmapping algorithms.
• Autonomous Navigation Stack: Localization using AMCL, global path planners (A*, Dijkstra), and local path planners (DWA, TEB).
• Hands-on: Simulating and building a Differential Drive Robot that maps an unknown environment and navigates autonomously to designated waypoints.

Phase 1: Omniverse & Scene Building

• Introduction & Interface Navigation: Overview of Omniverse vs. Gazebo, system requirements, launch process, and navigating the GUI.
• USD (Universal Scene Description) Core Concepts: Prims, Transforms, Layers, and References.
• Physics Setup & Colliders: Rigid Body dynamics, Colliders, mass, friction, and Physics Materials.
• Environment Creation: Building a basic workspace using ground planes, lights, wall obstacles, and CAD/USD props.
• Hands-on Practice & Guided Lab: Create and save a USD warehouse/obstacle environment from scratch.

Phase 2: Robot Import & Articulation

• URDF Importer: Importing a custom robot URDF into Isaac Sim.
• Robot Articulation & Joints: Configuring Articulation Roots, Joint Drives, and manual joint controls.
• Adding Camera Sensors: Attaching RGB and Depth cameras to the robot frame.
• Adding Distance Sensors (LiDAR): Attaching RTX 2D/3D LiDAR sensors.
• Hands-on Practice & Guided Lab: Import a differential-drive robot or arm, attach LiDAR and Camera, and drive it manually.

Phase 3: ROS 2 Bridge & Interactive Control

• Introduction to Action Graph & ROS 2 Bridge: Enabling the ROS 2 Extension, OmniGraph nodes, clock synchronization, and TF publishing.
• Driving the Robot via ROS 2: Building an Action Graph to convert incoming ROS 2 Twist messages into Isaac Sim differential-drive joint velocities.
• Publishing Sensor Feeds to ROS 2: Routing Isaac Sim LiDAR data to /scan and camera frames to /image_raw.
• End-to-End Simulation Test: Running ros2 run teleop_twist_keyboard to drive the Isaac Sim robot while viewing the /scan topics inside RViz2.
• Mini-Project & Buffer Day: Teleoperate the robot car through an obstacle course in Isaac Sim.

Full-System Robotics Integration

• System Architecture: Unifying physical hardware, microcontroller firmware, ROS middleware, and AI vision models.
• Development Lifecycle: Hardware assembly, power management, software stack deployment, integration testing, and documentation.
• Industry-oriented capstone project integrating the complete robotics stack.
• Internship experience with practical robotics development workflows.