Generalizations to Control Laws of Image-Based Visual Servoing

In this article, we review the existing control laws used in the image-based visual servoing (IBVS). In particular, we discuss the issues related to the conventional interaction-matrix approximations for IBVS, namely accuracy of approximations, full-rankedness of the interaction matrix, the camera retreat problem, and the stability of the resultant control laws. In order to address some of these issues, we propose parametric approximations for the interaction matrix and formulate new control laws accordingly. In particular, we show that other control methods are special cases of our generalized control method. We show the features and advantages of our methods in terms of the interaction matrix rank-deficiency, immunity to camera retreat problem, and local stability. Simulation results are presented to support the advantages of our proposed techniques.