Image-based visual servo control using the port-Hamiltonian approach

Abstract

This work is devoted to an image-based visual servo control strategy for standard mechanical systems in the port-Hamiltonian framework. We utilize a change of variables that transforms the port-Hamiltonian system into one with constant mass-inertia matrix, and we use an interaction matrix that includes the depth information together with the image features variables of the image plane. We develop a control strategy that renders a closed-loop system that is port-Hamiltonian. The introduced approach is applied to a two degrees of freedom robot arm problem, and simulation results are provided.