Evitement d’obstacles pour un robot dans un environnement TR

Abstract

In the real world, many systems are characterized by complex nonlinear dynamics behavior. Control of mobile robot classified in this class. Currently, there is no systematic theory and generally applicable to control such systems. Favor of their universal approximation capability, techniques of artificial intelligence based on fuzzy logic and neural network are considered as a very interesting solution to address these problems. The use of these techniques for autonomous navigation with obstacle avoidance of a mobile robot has expanded significantly in recent years. In this paper we discuss a strategy to integrate these techniques into the framework of a cooperative system of navigation. We try to provide the capabilities offered by these techniques and their limitations, through a simulation experiment.