Learning behavioral control by reinforcement for an autonomous mobile robot

摘要

We present an implementation of a reinforcement learning algorithm through the use of a special neural network topology, the AHC (adaptive heuristic critic). The AHC constitutes a fusion supervisor of primitive behaviours in order to execute more complex robot behaviours as for example go to goal. This fusion supervisor is part of an architecture for the execution of mobile robot tasks which are composed of several primitive behaviours which act in a simultaneous or concurrent fashion. The architecture allows for learning to take place at the execution level, it incorporates the experience gained in executing primitive behaviours as well as the overall task. The implementation of the autonomous learning approach has been tested within OPMOR, a simulation environment for mobile robots and with our mobile platform UPM Robuter. Both simulated and real results are presented. The performance of the AHC neural network is adequate. Portions of this work have been implemented in the EEC ESPRIT 2483 PANORAMA Project.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>