Robot shaping: The Hamster Experiment

Abstract

. In this paper we present an example of the application of a technique, which we call robot shaping, to designing and building learning autonomous robots. Our autonomous robot (called HAMSTER 1 ) is a multi-sensor mobile robot that performs the task of collecting "food" and bringing it to its "nest". Its control architecture is based on the behavioral paradigm. The behavioral modules are implemented as classifier systems and are learned by a reinforcement learning technique exploiting the Bucket Brigade and an extended version of the Genetic Algorithm. The chief features of HAMSTER are that it combines innate (i.e., prewired) and learned behaviors, and that training was carried out in a simulated environment and then transferred to the real robot. Published in Proceedings of ISRAM'96, Sixth International Symposium on Robotics and Manufacturing, M.Jamshidi et al. (Eds.), May 28--30, 1996, Montpellier, France. ^ Currently Ph.D. student at the University of Padua, Italy. 1 HAMSTER :...