Abstract
This paper investigates visual feedback pose (position and attitude) synchronization in leader-following type visibility structures on the Special Euclidean group SE(3). We first define visual robotic networks for multiple camera models. We then propose a visual feedback pose synchronization law combining a vision-based observer with the pose synchronization law presented in our previous works. We then prove that for a static leader, the present control law on the network achieves visual feedback pose synchronization. Moreover, for a moving leader, we evaluate the tracking performance of the network by employing the notion of L2-gain performance. Finally, the validity of the present control law and analysis is demonstrated through experiments.
