The movement of mobile robots involves challenges that do not appear in stationary manipulators. Some of these challenges include noisy measurements and inaccurate estimations of the robot’s pose. These factors lead to uncertainties in the robot’s motion, which may entail inappropriate control actions or even system instability. For that reason, this study of techniques for managing such uncertainty is a key piece of research within the mobile robotics field.

This book is about approaches to robotic perception using the concepts of probability theory. It is true that this book deals with just a part of the contents of Thrun et al.’s well-known book [1], but it is important to note that it covers in more detail the issue of uncertainty in robotic perception, or how to obtain meaningful information from sensory raw data. In particular, it addresses the fundamentals of Bayesian inference and related concepts to discuss modeling robotic perception, decision and control, and learning systems. After that, the authors apply the proposed methodology to a couple of case studies in real environments. The book concludes with an evaluative discussion of future trends in comparison to other techniques.

This book is quite readable and includes many figures that aid in understanding the theoretical concepts. From the very beginning, the authors clearly state the contents and intended audience. This text gains extra value from an accompanying website with supplementary material.