Introduction

This introduction gives an overview of embodied AI: what it studies, where it came from, which skills and body types robots have, and which problems remain open. The final two chapters explain how to learn the technical modules and how career roles divide the work. After the introduction, move on to the foundations and projects to learn specific techniques.

Where the introduction fits

Section	What it does	When to read it
Introduction	Helps beginners get started: what the field is, why it is hard, and which approaches exist	On your first visit, to build an overall understanding
Foundations	Explains the principles: reinforcement learning, VLA, control, perception, simulation, and more	When you hit a specific knowledge gap
Projects	Offers projects organized into chapters, standalone experiments, demos, and reproductions	When you want to build a complete system or test a method yourself

Chapters

Chapter	Questions it answers
1. What is embodied AI	What does embodied AI study? How does it differ from AI that only processes text and images?
2. A brief history	Which periods has embodied AI gone through, and where did robot capabilities come from in each?
3. Robot skills	How do you define a robot task? What are grasping, manipulation, locomotion, and navigation?
4. Embodied AI platforms	What are the common robot bodies? How does the body shape which experiments are possible?
5. Key challenges	Why are data, sim-to-real transfer, generalization, real-time execution, safety, and evaluation hard?
6. Technology stack	What does each learning module do, and which theory courses and projects should you start with?
7. Career roles	What does each role do, which skills does it need, and how can you prepare for interviews?

Resources

The resources collect papers, datasets, open source projects, and simulation tools. Each entry links to its original source and related learning material on this site.

Category	What you can find
Papers and research	Research questions, original papers, project pages, and our explanations
Embodied AI datasets	Robot demonstrations, task types, and evaluation benchmarks
Open source projects	Model code, training frameworks, and reproduction entry points
Simulation and tools	Simulation engines, learning environments, and official documentation

How to read

• New to embodied AI: read chapters 1 → 3 → 4, then go to projects to choose your first experiment.
• Already familiar with machine learning: read chapters 1 → 2 → 5 for an overview of the field's history and open problems.
• Starting structured learning: read 6. Technology stack, then follow its modules into theory courses and projects.
• Exploring career directions: read 7. Career roles for responsibilities and preparation topics.
• Looking for papers, data, or code: go straight to the resources.

The introduction requires no programming or math background. Each chapter ends with further reading on this site for when you want to go deeper.