Probabilistic Robotics by Sebastian Thrun, Wolfram Burgard and Dieter Fox introduces a wealth of techniques and algorithms in the field. All algorithms are primarily based on a single overarching mathematical foundation. Constructing on the sphere of mathematical statistics, probabilistic robotics endows robots with a brand new degree of robustness in actual-world situations.
Probabilistic robotics is a new and rising area in robotics, concerned with perception and management within the face of uncertainty. Every chapter offers instance implementations in pseudo code, detailed mathematical derivations, discussion from practitioner’s perspective, and intensive lists of workouts and sophistication projects.
The book is relevant for anybody concerned in robotic software program development and scientific research. It’ll also be of interest to applied statisticians and engineers coping with real-world sensor data. It shows three examples of profitable robotic programs working in unsure environments: a commercially deployed autonomous straddle carrier, an interactive museum tour-guide robotic and a prototype robotic assistant.
The probabilistic method addresses them by way of a single concept: methods to symbolize information probabilisticly. Particularly, world fashions within the probabilistic method are conditional probability distributions describing the dependence of sure variables on different variables in probabilistic terms. A robotic’s state of information can also be represented by probability distributions derived by integrating sensor measurements into the probabilistic world fashions given to the robot.
Probabilistic Robotics (Intelligent Robotics and Autonomous Agents series) [Hardcover]
Sebastian Thrun, Wolfram Burgard and Dieter Fox
The MIT Press (August 19, 2005)