Week-by-Week Course Timeline
Overview
This course is structured as a 13-week intensive program designed to take you from fundamentals to building autonomous humanoid robots.
Week 1-2: Foundations
Week 1: Introduction to Physical AI
- Topics:
- Digital vs Embodied Intelligence
- Robotics ecosystem overview
- Development environment setup
- Lab: Install ROS 2 Humble, Python 3.10, Gazebo Classic
- Assignment: Environment verification report
Week 2: ROS 2 Fundamentals
- Topics:
- Nodes, Topics, Services
- Publisher/Subscriber patterns
- ROS 2 workspace structure
- Lab: Build your first ROS 2 node
- Assignment: Multi-node communication system
Week 3-5: Module 1 - The Robotic Nervous System
Week 3: Python + ROS 2 Integration
- Topics:
- rclpy library deep dive
- Custom messages and services
- Parameter management
- Lab: Python-controlled robot arm simulation
- Assignment: Voice-to-motion controller
Week 4: Robot Description (URDF)
- Topics:
- URDF syntax and structure
- Joints, Links, and Kinematics
- Xacro for modular robot definitions
- Lab: Model a humanoid torso in URDF
- Assignment: Full humanoid URDF with 20+ joints
Week 5: ROS 2 Mini-Project
- Project: Design a robotic nervous system that responds to sensor inputs
- Deliverable: GitHub repository with documentation
Week 6-8: Module 2 - Digital Twin & Simulation
Week 6: Gazebo Physics Engine
- Topics:
- World files, SDF format
- Physics parameters (gravity, friction, inertia)
- Sensor plugins (cameras, LIDAR, IMU)
- Lab: Simulate a humanoid in a physics environment
- Assignment: Custom world with obstacles
Week 7: Advanced Sensors & Unity Integration
- Topics:
- Depth cameras, force-torque sensors
- Unity-ROS bridge
- High-fidelity visualization
- Lab: LIDAR-based mapping in Gazebo
- Assignment: Unity visualization of robot telemetry
Week 8: Sim-to-Real Gap
- Topics:
- Domain randomization
- Sensor noise modeling
- Calibration strategies
- Lab: Compare sim vs real sensor data
- Assignment: Analysis report on sim-to-real differences
Week 9-11: Module 3 - The AI-Robot Brain
Week 9: NVIDIA Isaac Sim
- Topics:
- Omniverse platform overview
- Isaac Sim workflow
- Synthetic data generation
- Lab: Train a perception model on synthetic data
- Assignment: Object detection pipeline
Week 10: Navigation & Path Planning
- Topics:
- Nav2 stack architecture
- VSLAM (Visual SLAM)
- Costmaps and obstacle avoidance
- Lab: Implement Nav2 for a mobile robot
- Assignment: Autonomous navigation in a complex environment
Week 11: Reinforcement Learning for Control
- Topics:
- RL basics (Q-learning, Policy Gradients)
- Isaac Gym for parallel simulation
- Training bipedal locomotion
- Lab: Train a walking policy with Isaac Gym
- Assignment: Evaluate policy on uneven terrain
Week 12: Module 4 - Vision-Language-Action (VLA)
Week 12: Conversational Robotics
- Topics:
- Speech recognition with Whisper
- LLM integration (GPT-4, LLaMA)
- Multi-modal fusion (text, voice, vision)
- Lab: Voice command system: "Pick up the red cube"
- Assignment: Build a VLA agent with memory
Week 13: Capstone Project
Final Week: Autonomous Humanoid System
- Objective: Build an end-to-end system
- Requirements:
- Voice command input (Whisper)
- LLM-based task planning
- Nav2 path planning
- Object detection and manipulation
- Obstacle avoidance
- Deliverable: Live demo + technical report
- Evaluation:
- System architecture (20%)
- Code quality (20%)
- Demo performance (30%)
- Documentation (20%)
- Innovation (10%)
Assessment Structure
| Component | Weight | Description |
|---|---|---|
| Weekly Labs | 30% | Hands-on exercises (Week 1-12) |
| Assignments | 30% | Take-home projects (5 total) |
| Mini-Projects | 20% | Module 1 & 2 milestones |
| Capstone | 20% | Final autonomous humanoid demo |
Prerequisites Check
Before Week 1, ensure you have:
- Ubuntu 22.04 or equivalent Linux environment
- Python 3.10+
- GPU with CUDA support (recommended: RTX 3060+)
- 16GB+ RAM, 50GB+ storage
- Git, Docker installed