Adaptive two layer fuzzy control of a mobile robot system

Abstract

Various approaches have been proposed for solving collision avoidance problems in robotic systems [l, 2, 3, 4, 5, 61. In this paper we propose a method for designing an adaptive two layer fuzzy control system for collision-free trajectory control of a multiple robot system. The proposed method is based on the integration of Genetic Algorithms and Fuzzy Logic. The robots are considered as point masses moving in a common work space. The knowledge base which controls how to navigate in the work space without collision is determined by learning via Genetic Algorithms.