Skip to main content

Physical AI Textbook

Learn Robotics with ROS2, Gazebo, Unity, and NVIDIA Isaac Sim

Start Learning →

Course Modules

Weeks 3-5

Module 1: The Robotic Nervous System (ROS 2)

Master ROS 2 architecture, communication patterns, and robot modeling. Learn to build distributed robotic systems using nodes, topics, services, and actions.

Learning Outcomes:
  • Explain the ROS 2 computation graph and its components
  • Create publishers, subscribers, and service clients using rclpy
  • Define robot structure using URDF and visualize in RViz2
Weeks 6-7

Module 2: Digital Twins - Simulation & Sensors

Build digital twins for robotic systems using Gazebo and Unity. Simulate sensors, physics, and environments for testing before deploying to physical hardware.

Learning Outcomes:
  • Create Gazebo simulation environments with physics and sensors
  • Integrate Unity for photorealistic sensor simulation
  • Test navigation and perception algorithms in simulation
Weeks 8-10

Module 3: NVIDIA Isaac - Perception & Navigation

Leverage NVIDIA Isaac Sim for GPU-accelerated robotics. Implement VSLAM, Nav2 navigation stacks, and reinforcement learning for autonomous behaviors.

Learning Outcomes:
  • Set up and configure NVIDIA Isaac Sim environments
  • Implement Visual SLAM for robot localization
  • Deploy Nav2 navigation stack for autonomous navigation
Weeks 11-13

Module 4: VLA & Humanoid Robotics

Integrate Vision-Language-Action models with humanoid robots. Master humanoid kinematics, manipulation, and conversational AI for natural human-robot interaction.

Learning Outcomes:
  • Calculate forward and inverse kinematics for humanoid robots
  • Implement manipulation primitives for pick-and-place tasks
  • Integrate conversational AI with robot action planning

Recent Updates

2025-11-29
Textbook Structure Initialized
Complete textbook structure with 4 modules and dashboard homepage is now live.
2025-11-29
Setup Guides Available
Hardware setup guides for Workstation, Edge Kit, and Cloud are now available.