Formation control based on adaptive NN with time-varying interaction among robots
Xin Chen, Min Wu, Yangmin Li · 2008
This paper addresses a formation control issue for a group of robots with dynamic interaction topology. Normally given a formation shape (pattern), if interaction topology changes from time to time, formation control developed for continuous systems may fail to keep formation, because the system may be not continuous any longer. This paper introduces a decentralized formation control to realize formation. Moreover based on analysis of switched systems, the sufficient conditions on formation pattern and interaction topology are fingered out to ensure the feasibility of this formation control, even if the interaction topology often changes suddenly. The simulation illustrates that within a field with obstacles, which may interrupt interaction between robots, a leader-following formation must reach the destination by using this formation control.