A Consensus-Based Approach for Visual Servo Control of Multiple Mobile Robots

This paper addresses a consensus-based approach for visual servo control of multiple mobile robots. The basic idea is to utilize single-integrator consensus techniques to design a image-based visual servo control algorithm for each robot under a certain network. The aim of this work is to extract common objects between different images captured by different robots and adjust the positions and orientations of robots to concentrate on common objects. We show that if the union of communication graphs has a directed spanning tree, the states of robots would achieve consensus based on the relative states. At the end of this paper we present experiments to verify the feasibility and effectiveness of proposed algorithms.