Skip to main content

Syllabus: Physical AI & Humanoid Robotics

Course Information​

Course Title: Physical AI & Humanoid Robotics Course Code: ROB 450/550 Credits: 4 Duration: 13 weeks Level: Undergraduate/Graduate Prerequisites: Basic programming skills, Linear Algebra, Calculus Instructor: [To be assigned] Semester: [To be determined]

Course Description​

This advanced course explores the intersection of artificial intelligence and robotics through the lens of Physical AI and embodied intelligence. Students will learn to design, implement, and deploy intelligent robotic systems that integrate perception, reasoning, and action in physical environments. The course emphasizes humanoid robotics, simulation-to-real transfer, and multi-modal AI systems that combine vision, language, and action.

Required Materials​

  1. Primary Textbook: This Physical AI & Humanoid Robotics Textbook
  2. Software: ROS 2 Humble, Isaac Sim, Gazebo, Unity Robotics
  3. Hardware: Access to robotic platforms (simulated and physical)
  4. Additional Resources: Research papers, online documentation, video tutorials

Four Core Modules​

Module 1: Physical AI and Embodied Intelligence (Weeks 1-3)​

Duration: 3 weeks Credits: 1 credit

Topics Covered:​

  • Foundations of Physical AI
  • Principles of Embodied Intelligence
  • ROS 2 as the Robotic Nervous System
  • Sensor Integration and Data Processing

Learning Objectives:​

  • Understand the theoretical foundations of Physical AI
  • Implement basic ROS 2 communication patterns
  • Integrate sensors into robotic systems
  • Process sensor data in real-time

Assessment:​

  • Weekly quizzes (20%)
  • ROS 2 implementation project (30%)
  • Sensor integration lab report (30%)
  • Participation and attendance (20%)

Resources:​

  • Chapters 1-3 of textbook
  • ROS 2 tutorials and documentation
  • Simulation environments

Module 2: Simulation and Perception Systems (Weeks 4-7)​

Duration: 4 weeks Credits: 1.5 credits

Topics Covered:​

  • Simulation environments (Gazebo, Isaac Sim, Unity)
  • AI perception and computer vision
  • SLAM and navigation systems
  • Isaac ROS perception pipelines

Learning Objectives:​

  • Create and configure simulation environments
  • Implement AI-based perception systems
  • Deploy SLAM and navigation algorithms
  • Optimize perception pipelines for performance

Assessment:​

  • Simulation environment project (35%)
  • Perception system implementation (35%)
  • Navigation system project (20%)
  • Module examination (10%)

Resources:​

  • Chapters 4-6 of textbook
  • Isaac Sim and Isaac ROS documentation
  • Gazebo tutorials
  • Unity robotics packages

Module 3: Humanoid Robotics and Control (Weeks 8-10)​

Duration: 3 weeks Credits: 1 credit

Topics Covered:​

  • Humanoid robot kinematics and dynamics
  • Locomotion control and balance
  • Humanoid robot architectures
  • Manipulation and interaction

Learning Objectives:​

  • Analyze humanoid robot kinematic structures
  • Implement locomotion and balance control
  • Design humanoid robot control architectures
  • Execute manipulation tasks on humanoid platforms

Assessment:​

  • Kinematics analysis project (25%)
  • Locomotion control implementation (35%)
  • Humanoid robot design project (25%)
  • Practical demonstration (15%)

Resources:​

  • Chapters 7-8 of textbook
  • Humanoid robotics documentation
  • Control theory resources
  • Simulation platforms with humanoid models

Module 4: Vision-Language-Action Systems (Weeks 11-13)​

Duration: 3 weeks Credits: 0.5 credits

Topics Covered:​

  • Vision-Language-Action (VLA) system architecture
  • Voice-command and natural language interfaces
  • Multi-modal AI integration
  • Capstone project integration and presentation

Learning Objectives:​

  • Design VLA system architectures
  • Integrate voice-command interfaces
  • Combine vision, language, and action systems
  • Implement and present capstone projects

Assessment:​

  • VLA system design (30%)
  • Voice-command interface implementation (25%)
  • Capstone project development (35%)
  • Final presentation and demonstration (10%)

Resources:​

  • Chapters 9-10 of textbook
  • Natural language processing resources
  • Voice recognition and processing tools
  • Multi-modal AI frameworks

Weekly Schedule​

Week 1: Introduction to Physical AI​

  • Course introduction and expectations
  • Overview of Physical AI and embodied intelligence
  • Basic concepts and historical context

Week 2: Robotics Foundations​

  • Kinematics and dynamics fundamentals
  • Sensors and actuators overview
  • Introduction to ROS 2 concepts

Week 3: ROS 2 - The Robotic Nervous System​

  • ROS 2 architecture deep dive
  • Nodes, topics, services, and actions
  • Parameter management and launch systems

Week 4: Simulation Environments - Gazebo​

  • Gazebo setup and configuration
  • Robot model creation and import
  • Physics simulation and sensor modeling

Week 5: Simulation Environments - Isaac Sim​

  • Isaac Sim architecture and capabilities
  • Photorealistic simulation
  • Isaac ROS integration

Week 6: Simulation Environments - Unity​

  • Unity robotics tools and packages
  • 3D environment creation
  • Real-time simulation and visualization

Week 7: AI Perception and Navigation​

  • Computer vision for robotics
  • SLAM algorithms and implementation
  • Path planning and navigation

Week 8: Humanoid Robotics - Kinematics​

  • Humanoid robot kinematics fundamentals
  • Forward and inverse kinematics
  • Jacobian and dynamic modeling

Week 9: Humanoid Robotics - Dynamics and Control​

  • Dynamic control principles
  • Balance and stability concepts
  • Walking pattern generation

Week 10: Humanoid Locomotion​

  • Bipedal locomotion principles
  • Control architectures for humanoid robots
  • Practical implementation and simulation

Week 11: Vision-Language-Action Systems​

  • VLA system architecture overview
  • Multi-modal AI concepts
  • Integration challenges and solutions

Week 12: Voice and Natural Language Interfaces​

  • Voice-command processing
  • Natural language understanding for robotics
  • Integration with robotic control systems

Week 13: Capstone Project Presentations​

  • Capstone project demonstrations
  • Peer evaluations and feedback
  • Course reflection and future directions

Assessment Methods​

Continuous Assessment (60%)​

  • Weekly quizzes and assignments: 15%
  • Lab reports and practical exercises: 20%
  • Module projects: 25%

Final Assessment (40%)​

  • Midterm examination (Modules 1-2): 15%
  • Capstone project: 20%
  • Final examination: 5%

Grading Scale​

  • A (93-100): Excellent - Demonstrates comprehensive understanding and exceptional application
  • A- (90-92): Outstanding - Shows strong understanding with minor areas for improvement
  • B+ (87-89): Very Good - Solid understanding with good application
  • B (83-86): Good - Adequate understanding with satisfactory application
  • B- (80-82): Satisfactory - Basic understanding with adequate application
  • C+ (77-79): Adequate - Limited understanding with minimal application
  • C (73-76): Passing - Minimal understanding with basic application
  • C- (70-72): Marginal - Deficient understanding with limited application
  • F (<70): Failing - Inadequate understanding and application

Learning Outcomes​

Upon successful completion of this course, students will be able to:

  1. Knowledge: Demonstrate comprehensive understanding of Physical AI and embodied intelligence principles
  2. Application: Design and implement integrated robotic systems combining perception, reasoning, and action
  3. Analysis: Evaluate the effectiveness of different simulation-to-real transfer methodologies
  4. Synthesis: Integrate multiple AI modalities (vision, language, action) into cohesive robotic systems
  5. Evaluation: Assess the performance and safety of humanoid robotic systems
  6. Creation: Develop novel robotic applications that demonstrate course concepts

Course Policies​

Attendance Policy​

Regular attendance is expected. More than two unexcused absences may result in grade reduction.

Late Assignment Policy​

Late assignments will be penalized by 10% per day unless prior arrangements are made.

Collaboration Policy​

Collaboration is encouraged for learning, but individual assignments must be completed independently. All sources must be properly cited.

Academic Integrity​

All work must be original. Any form of academic dishonesty will result in appropriate disciplinary action.

Accommodation Policy​

Students with documented disabilities should contact the Office of Disability Services to arrange appropriate accommodations.

Technology Requirements​

Students must have access to:

  • Computer with sufficient specifications for simulation environments
  • High-speed internet connection
  • ROS 2 compatible operating system (Ubuntu 22.04 recommended)
  • Development tools (IDE, Git, etc.)

Support Resources​

  • Office hours: [Schedule to be announced]
  • Teaching assistants: [Contact information to be provided]
  • Online discussion forum
  • Tutorial sessions
  • Academic support services