Adaptive regulation of monocular camera-based mobile robots

For a monocular camera-based mobile robot system, an adaptive visual regulation controller is proposed in this paper to asymptotically stabilize the robot to its desired position/orientation, in which range sensor is not required and unknown camera extrinsic parameters are considered as well. First, based on the relationship between the robot frame and the camera frame, kinematic model of the monocular mobile robot system is established with respect to the camera coordinate system. Then, based on the model, observable error signals are extracted from image by decomposing the homography matrix and introducing a proper coordinate transformation, which are subsequently employed to construct the proposed adaptive stabilization controller based on Lyapunov techniques. The performance of the controller is demonstrated by both theoretical stability analysis and simulation results, which illustrate that the proposed adaptive controller can achieve the regulation task asymptotically in presence of unknown camera extrinsic parameters.