Weeks 1-2: Introduction to Physical AI

Learning Objectives

During these first two weeks, students will establish a foundational understanding of Physical AI concepts and the broader field of embodied intelligence.

Topics Covered

• Foundations of Physical AI and embodied intelligence

  • Understanding the difference between digital AI and physical AI
  • The importance of embodiment in AI systems
  • How physical interaction shapes intelligence

• From digital AI to robots that understand physical laws

  • The transition from simulation to reality
  • Physical constraints and considerations
  • Real-world sensor and actuator limitations

• Overview of humanoid robotics landscape

  • Current state of humanoid robotics
  • Key players and technologies
  • Future directions and applications

• Sensor systems: LIDAR, cameras, IMUs, force/torque sensors

  • Types of sensors used in robotics
  • Sensor fusion concepts
  • Data acquisition and processing

Key Concepts

Physical AI vs. Digital AI

Physical AI represents a fundamental shift from AI systems that operate purely in digital spaces to systems that function in and interact with the physical world. This introduces new challenges:

• Real-time constraints: Physical systems have timing requirements
• Uncertainty: Real-world sensing is noisy and incomplete
• Safety: Physical systems can cause damage if not properly controlled
• Embodiment: The physical form affects the AI's capabilities and constraints

Embodied Intelligence

Embodied intelligence is the theory that intelligence emerges from the interaction between an agent and its environment. Key principles include:

• The body shapes cognition
• Intelligence is distributed between brain, body, and environment
• Physical interaction is essential for certain types of learning

Assignments

1. Research paper review: "The Importance of Embodied Intelligence"
2. Setup development environment for the course
3. Basic ROS 2 installation and configuration