Part Overview
This part covers memory, continual learning, open problems, capstone projects, and teaching paths. It connects formal ideas with the tools and labs needed to build working systems.
Chapters: 5. Each chapter includes theory, recipes, practical code, a library shortcut, and exercises.
Frontiers, Capstones, and Course Design gives the reader a working layer of the embodied AI stack. Later chapters assume this layer when agents must perceive, plan, act, and recover from mistakes.
This chapter develops embodied agents with memory as part of the embodied AI stack.
- 56.1 Why memory matters; short- vs. long-term
- 56.2 Spatial, episodic, and semantic memory
- 56.3 Memory retrieval for planning
- 56.4 Memory errors
This chapter develops continual and lifelong learning as part of the embodied AI stack.
- 57.1 Learning after deployment
- 57.2 Catastrophic forgetting and mitigation
- 57.3 Online adaptation; human correction as data
- 57.4 Safe continual learning; evaluation over time
This chapter develops frontier and open problems as part of the embodied AI stack.
- 58.1 Scaling laws and data engines for robots
- 58.2 Generalist vs. specialist policies
- 58.3 World models in the robot loop
- 58.4 The open-vs-closed model divide
- 58.5 What is still unsolved (long-horizon reasoning, reliability, real-world RL)
- 58.6 Frontier Watch
This chapter develops capstone projects as part of the embodied AI stack.
- 59.1 Object search in a simulated home
- 59.2 Language-guided navigation with replanning
- 59.3 Vision-based robotic pick-and-place (IL + RL)
- 59.4 Fine-tune an open VLA on a custom task (LeRobot)
- 59.5 Learned locomotion with sim-to-real analysis
- 59.6 World-model-based planning agent
- 59.7 Safety-shielded embodied agent
- 59.8 LLM-based household task planner
- 59.9 Drone inspection planner
- 59.10 Multi-agent search and rescue
- 59.11 Open-ended research project
- 59.12 Application track capstone templates
This chapter develops teaching with this book as part of the embodied AI stack.
- 60.1 One-semester graduate course (14 weeks)
- 60.2 One-semester advanced undergraduate course (lighter theory, more labs)
- 60.3 Two-semester sequence
- 60.4 Research-seminar track
- 60.5 Lab infrastructure and compute budgeting for instructors
- 60.6 Assessment, rubrics, and academic-integrity notes for code assignments
Part XII Production Frame
Part XII turns the book from a sequence of mechanisms into a set of durable practices. Memory chapters ask what past evidence should be retrieved. Continual-learning chapters ask how deployed systems change without erasing what already works. Frontier chapters ask which claims survive reproducible evaluation. Capstone and teaching chapters convert the whole book into projects, labs, rubrics, and course paths.
| Chapter | Primary Deliverable | Audit Question |
|---|---|---|
| 56 Memory | Memory contract and retrieval test | Which retrieved evidence changed the action? |
| 57 Continual Learning | Versioned update panel | Did the update improve the new task without erasing protected skills? |
| 58 Frontiers | Frontier claim watchlist | Which claims have artifacts, independent tests, or reproducible protocols? |
| 59 Capstones | Portfolio-grade project folder | Can the result be rerun, critiqued, and improved? |
| 60 Teaching | Undergraduate, graduate, and seminar path | Does each week end with an artifact and an evidence discussion? |
What's Next?
After this part, the appendices consolidate tools and references.