Fuzzy-neuro control design for obstacle avoidance of a mobile robot

We present a new method for real-time obstacle avoidance of a mobile robot based on a structure of fuzzy-neuro network. The design goal was to facilitate a mobile robot to reach the target safely and efficiently in a completely unknown environment. The inputs of the network are quantitized ultrasonic sensory data. The outputs are velocity commands to motor controllers. Useful heuristic rules are combined with a neural network computation architecture for building the desired mapping between perception and motion. The method is more efficient and less sensitive to sensor misreadings compared with the conventional approaches. It allows continuous and fast motion of the mobile robot without stopping by obstacles. The effectiveness of the proposed method is demonstrated by a series of practical tests on our experimental mobile robot.