BEHAVIORS FUSION FOR A MOBILE ROBOT USING FUZZY BEHAVIORS WEIGHTS TUNED BY REAL CODED GENETIC ALGORITHM

Coordination between different behaviors, such as goal reaching, obstacle avoidance, reaching the collision–free space, and wall following, all of these represent a big problem in navigation of the mobile robots in unknown or partially known environments. In this paper a solution of this problem is presented using switched gains, these gains are inferred from a fuzzy fusion system (FFS). These gains are weighted and tuned by a real coded Genetic Algorithm (RCGA) based on a fitness function that ensures a safe and short path to the object. The proposed technique assures a safe and efficient navigation.