Adaptive Iterative learning control for multi-agent systems consensus tracking

Abstract

This paper addresses an adaptive iterative learning control (AILC) based scheme for multi-agent systems (MAS) consensus tracking under repeatable control environment. The agent dynamics are assumed to be inherently nonlinear with unknown time-varying parameters. The underline communication among followers is fixed and undirected. The leader's trajectory is dynamically changing, and only available to a small portion of followers. By utilizing the repetitiveness of the tracking task, the unknown time-varying parameters can be effectively estimated, and the leader's velocity is not required in the controller. Under either resetting condition or alignment condition, perfect consensus tracking for the MAS can be achieved asymptotically in the iteration domain. A simulation example is given to demonstrate the effectiveness of the proposed algorithm. © 2012 IEEE.