Fuzzy Navigation Of A Mobile Robot

Abstract

A navigation system based on fuzzy logic controllers is developed for a mobile robot in an unknown environment. The structure of this fuzzy navigation system features the com- bination of sensor system, fuzzy controllers for motion planning and the motion control system for real-time execution. Six ultrasonic sensors on-board the mobile robot are used for distance measurement to the immediate obstacles. Sensor data are fuzzified to be the inputs of the fuzzy controller. Three states, each with five quantized levels are used to define the fuzzy set. Two fuzzy controllers are designed to handle the naviga- tion problem. Each fuzzy controller, which cor- responds to the turn right or turn left condition, has four inputs, two outputs and 81 rules. The outputs are the command velocities to the left and right wheels, which drive the mobile robot. These command velocities are sent to the lower level motion control system. The performance of this navigation system is tested by computer sim- ulation. Satisfactory results have been obtained and are presented in this paper.