The problem in the autonomous navigation of a mobile robot is to define a strategy that allows it to reach the final destination and avoiding obstacles. Fuzzy logic is considered as an important tool to solve this problem. It can mimic reasoning abilities of the human being in navigation tasks. However a major problem of fuzzy systems is obtaining their parameters which are generally specified by human experts. This process can be long and complex. In order to generate optimal parameters of fuzzy controller, this work propose a learning and optimization process based on ant colony algorithm ACO and genetic algorithm operators (crossover and mutation).We present a comparison between inference system for autonomous navigation based on fuzzy logic before and after learning. The simulated results show clearly the impact of the optimization approach improves the fuzzy controller performance mainly in obstacle avoidance and detection of the shortest path.

This paper proposes a hybrid approach based on limit-cycles method and fuzzy logic controller for the problem of obstacle avoidance of mobile robots in unknown environment. The purpose of hybridization consists on the improvement of basic limit-cycle method in order to obtain safe and flexible navigation. The proposed algorithm has been successfully tested in different configurations on simulation.