Robot Shaping: An Experiment in Behavior Engineering
ISBN: , SKU: , AUTHOR: Dorigo, Marco / Colombetti, Marco, PUBLISHER: Bradford Book, foreword by Lashon Booker To program an autonomous robot to act reliably in a dynamic environment is a complex task. ong>Theong> dynamics ong>ofong> ong>theong> environment are unpredictable, and ong>theong> robots' sensors provide noisy input. A learning autonomous robot, one that can acquire knowledge through interaction with its environment and ong>theong>n adapt its behavior, greatly simplifies ong>theong> designer's work. A learning robot need not be given all ong>ofong> ong>theong> details ong>ofong> its environment, and its sensors and actuators need not be finely tuned. "Robot Shaping" is about designing and building learning autonomous robots. ong>Theong> term "shaping" comes from experimental psychology, where it describes ong>theong> incremental training ong>ofong> animals. ong>Theong> authors propose a new engineering discipline, "behavior engineering," to provide ong>theong> methodologies and tools for creating autonomous robots. ong>Theong>ir techniques are based on classifier systems, a reinforcement learning architecture originated by John Holland, to which ong>theong>y have added several new ideas, such as "mutespec," classifier system "energy," and dynamic population size. In ong>theong> book ong>theong>y present Behavior Analysis and Training (BAT) as an example ong>ofong> a behavior engineering methodology.