A vision-based navigation control system for a mobile service robot

In this paper, we describe the development of a control system for autonomous navigation of a mobile service robot. Visual information from a single color CCD camera are utilized together with range sensors to perceive the robot dynamic environment. The target is extracted from the image by its known color, and its properties are then measured and fed to the controller. A behavior-based control strategy is proposed which incorporates two basic behaviors of target tracking and obstacle avoidance. Fuzzy logic control, FLC, is used for behavior design to tolerate effectively the uncertainties associated with observations. Potential field theory is used for blending the two behaviors outputs into final decisions for robot control. The performance of the proposed system is demonstrated by real experiments on the robot. Experiments showed that the robot, endowed with our control system, can follow a colored target and avoid obstacles during its course of motion successfully.