A Fuzzy Agents Architecture for Autonomous Mobile Robots

Abstract

This paper describes the work we are carrying on with intelligent agents for mobile autonomous robots. We have developed an agents architecture, based on the blackboard paradigm, which follows theses guidelines: flexibility, simplicity, and fault tolerance. These objectives are accomplished by means of an extensive use of fuzzy logic at all software levels: agents behaviour definition, agents result fusion, sensors filtering, and actuators control. In addition to this, the architecture is intended to be used with learning techniques, to create or tune the fuzzy rule bases, and specially the set of fuzzy metarules. Some results are presented here, that have been obtained with both a low-end custom robot we have built and a simulator we have developed. 1 Introduction The operation of a mobile autonomous robot in an unstructured environment, as it occurs in the real world, needs to take into account many details. Mainly, the controller has to be able to operate under conditions of impre...