Agentic AI & Physical AI: Build Autonomous Systems with ROS 2, Reinforcement Learning, Sensor Fusion, and Sim-to-Real Transfer

Open-Source Textbook: Physical AI & Humanoid Robotics

This is an interactive, open-source textbook designed to guide learners through the entire pipeline of building autonomous humanoid robots, from foundational concepts to a final capstone project.

• Structure: The curriculum is organized into 7 Parts and 22 Chapters featuring 87 interactive lessons. It aligns perfectly with your appendices, including a glossary, mathematical foundations, and hardware/software guides.

• Content Focus:

  • Part 1 (Foundations): Bridges the gap between digital AI and physical systems.

  • Part 2-4 (Tools): Provides in-depth training on ROS 2, simulation environments like Gazebo and NVIDIA Isaac Sim, and the NVIDIA Isaac platform.

  • Part 5 (Humanoid Development): Covers advanced topics like balance, inverse kinematics, and whole-body control.

  • Part 6-7 (Integration): Explores conversational AI, vision-language-action models, and a hands-on capstone project.

• Key Features: It's an "AI-native" project, optimized for human learners and AI agents, and includes hands-on code examples ready to run in a browser without any installation.

Academic Textbook: The Principles and Practice of Embodied Intelligence

This is an academic volume published by Springer that offers a structured, systematic overview of the technical approaches in embodied intelligence, written by researchers from Sun Yat-sen University.

• Structure: The book is divided into 8 chapters, progressing from basic concepts to advanced methods.

• Content Focus:

  • Core Technologies: It begins with foundational tech like 3D spatial representation and reinforcement learning.

  • Key Applications: Chapters systematically cover essential embodied AI technologies, including perception, navigation, manipulation, planning, and multi-agent collaboration.

  • Practical Implementation: The final chapter is dedicated to implementing these methods on simulation platforms.

Edited Volume: Building Embodied AI Systems

This book, part of Springer's "Information Systems Engineering and Management" series, takes a broader, application-oriented look at embodied AI systems across various industries.

• Content Focus:

  • Real-World Applications: It explains how embodied systems like robots and drones are used to create "smarter environments," such as hospitals, warehouses, retail stores, and defense zones.

  • Core Technologies: It includes chapters on the technology behind these systems, as well as covering key challenges, ethics, and specific applications like autonomous driving and healthcare.

These resources together provide a comprehensive blend of theoretical foundations, hands-on technical training, and real-world application, mirroring the structure of your appendices.