Monitoring-Based Visual Servoing Of Wheeled Mobile Robots

In this paper, a novel monitoring-based visual servoing method is proposed with respect to monocular wheeled mobile robot systems, which can complete the stabilization task under a dynamic environment. For existing methods, the mobile robot can reach the desired pose correctly with stationary visual targets, but they cannot be applied when visual targets are moved due to external disturbance or purposeful manual operations. To deal with this problem, a fixed camera is introduced to monitor the movement of visual features, so that the proposed approach can still work well with dynamic targets. Firstly, utilizing relevant images captured by the monitor camera, the motion information of visual targets can be estimated by employing the POSIT algorithm. Then, the relationship between relevant coordinate systems and feature points is calculated. With real-time image feedback, the rotation matrix and translation vector between current and desired poses of the mobile robot can be obtained via coordinate system transformation. Finally, under polar coordinate representation, a motion controller is adopted to drive the robot to the desired pose. Both simulation and experimental results are collected to verify the feasibility of the proposed approach.