Fuzzy Control of a Mobile Robotic Vehicle

Abstract

This paper presents a fuzzy logic control scheme for the navigation of a mobile robot in the presence of obstacles. A fuzzy navigation controller is described which guides the robot from a start position to a goal, or sub-goal, position assuming that no obstacles are in the path. Obstacles affect the navigation controller according to a set of fuzzy inhibitive rules that take into account the vehicle geometry, the distance of the obstacle from the robot and the probability of the object being at the position indicated. To ensure that all possible collision-free paths are considered, each entry in the fuzzy fit vector is distributed, or spread, across the output universe of discourse before using a sliding window defuzzification technique to produce a crisp output value. The use of the sliding window defuzzification technique helps to remove indecision from the controller. Both simulated and laboratory test results are presented.