Cooperative pointing control of multi-robot systems via distributed target estimation

Abstract

In this paper, we propose a novel cooperative pointing control framework for multi-robot systems subject to one target. The existing approach exploits the geometric relations among neighboring robots and the target to design cooperative pointing controllers, which involves intricate geometric analysis and hinders us from applications in complex environments. Herein, all the robots and the target are stationary within the same plane, and their deployment is arbitrary. Besides, each robot knows its own global position. Without loss of generality, we assume that, two leader robots already point at the target and the three of them are not collinear. This setup in fact determines the target’s position implicitly, which but is unknown to the two leader robots as well as others. We thus design a distributed estimator to yield an asymptotic estimate of the target’s position at each robot, which gives the asymptotic estimate of the desired orientation angle for each non-leader robot. However, the estimate of the desired orientation angle follows specific geometric relations from the estimate of the target’s position that cannot be equal to the position of the corresponding robot. A state-dependent switching strategy is employed to enable the distributed target estimator to stay away from this singularity. Through tracking the estimate of the desired orientation angle, each non-leader robot regulates its own orientation angle to point to the target asymptotically. Finally, numerical simulations are presented to illustrate the validity of this framework.