Learning for Adaptive and Reactive Robot Control A Dynamical Systems Approach

Methods by which robots can learn control laws that enable real-time reactivity using dynamical systems; with applications and exercises.

This book presents a wealth of machine learning techniques to make the control of robots more flexible and safe when interacting with humans. It introduces a set of control laws that enable reactivity using dynamical systems, a widely used method for solving motion-planning problems in robotics. These control approaches can replan in milliseconds to adapt to new environmental constraints and offer safe and compliant control of forces in contact. The techniques offer theoretical advantages, including convergence to a goal, non-penetration of obstacles, and passivity. The coverage of learning begins with low-level control parameters and progresses to higher-level competencies composed of combinations of skills. 
 

Learning for Adaptive and Reactive Robot Control is designed for graduate-level courses in robotics,  with chapters that proceed from fundamentals to more advanced content. Techniques covered include learning from demonstration, optimization, and reinforcement learning, and using dynamical systems in learning control laws, trajectory planning, and methods for compliant and force control . 
Features for teaching in each chapter: 
 

applications, which range from arm manipulators to whole-body control of humanoid robots;

pencil-and-paper and programming exercises;

lecture videos, slides, and MATLAB code examples available on the author’s website . 

an eTextbook platform website offering protected material[EPS2]  for instructors including solutions.

 
 

Aude Billard is Professor, School of Engineering, Ecole Polytechnique Federale de Lausanne (EPFL) and Director of the Learning Algorithms and Systems Laboratory (LASA). Sina Mirrazavi is a Senior Researcher at Sony. Nadia Figueroa is the Shalini and Rajeev Misra Presidential Assistant Professor in the Mechanical Engineering and Applied Mechanics (MEAM) Department at the University of Pennsylvania.

Preface xiii
Notation xix
I Introduction 1
1 Using and Learning Dynamical Systems for Robot Control--Overview 3
2 Gathering Data for Learning 27
II Learning a Controller 43
3 Learning a Control Law 45
4 Learning Multiple Control Laws 111
5 Learning Sequences of Control Laws 131
III Coupling and Modulating Controllers 173
6 Coupling and Synchronizing Controllers 175
7 Reaching for and Adapting to Moving Objects 195
8 Adapting and Modulating an Existing Control Law 219
9 Obstacle Avoidance 245
IV Compliant and Force Control with Dynamical Systems 267
10 Compliant Control 269
11 Force Control 295
12 Conclusion and Outlook 303
V Appendices 
A Background on Dynamical Systems Theory 307
B Background on Machine Learning 315
C Background on Robot Control 357
D Proofs and Derivations 361
Notes 379
Bibliography 383
Index 391