Vision-Based Adaptive Impedance Control for Robotic Polishing

The wall polishing is a common and important task for the interior renovation. However, the current procedure of wall polishing is heavily dependent on manual works, which have several issues on safety, efficiency, and quality. In this paper, a new vision-based adaptive impedance controller is proposed for polishing robots to automate the labor-intensive works and hence systematically deal with the problems associated with manual polishing. The contributions of this paper are twofold. First, the vision feedback is employed in the desired impedance model such that the high resolution of the camera guarantees the quality of polishing. Second, the convergence to the desired impedance model is theoretically proved with the consideration of the online adaptation of uncertain dynamic parameters. Experimental results on a polishing robot are presented to illustrate the performance of the proposed controller.