Vision-Based Position Regulation of Differential-Drive Mobile Robots

This paper considers the problem of vision-based stabilization (a stopping maneuver) of nonholonomic differential-drive mobile robots. First, a camera-robot model suitable for control applications is developed. Using a look-ahead approach, a vision-based feedback controller is proposed, which stabilizes the robot at a desired position indicated by a static circular marker on the line to follow. An algorithm for circle detection based on Hough transform is presented for determining the pose of the robot with respect to a detected stopping landmark. Simulation and experimental results illustrate the effectiveness of the proposed controller.