A Fuzzy Logical Application in a Robot Self Navigation

Abstract

Wheel speed and turn angle are the main parameters in the self navigation for a robot. Since the parameters of self navigation significantly influence robot walking time and robot walking quality, techniques of determining optimal route parameters are in high demand in robot industry. Usually the turn angle is determined and fixed according to certain environment before robot walking; and the wheel speed is determined subjectively by either by robot operators or programmers. As a result, the wheel speed and turn angle is not optimal in terms of the real condition at every corn location, and it is fixed at a conservative value causing longer robot walking time. In this work, a generic and intelligent approach of robot path determination is proposed. First a simplified robot navigation model is introduced. Second a fuzzy rule-based system is established to control robot wheel speed based on the distance from the obstacles, and the turn angle. Then the geometric features of the environment are identified, the wheel speed is determinate and the optimal route is found. Finally this approach is applied on an example for robot navigation, and the results are simulated with computer system to demonstrate the advantages of this approach.