Transient behavior of synchronized agents: A design method for dynamic networked controllers

Abstract

The main objective of the synchronization of multi-agent systems is the asymptotic convergence of the agents towards a common trajectory. For practical purposes, it is required that the transient behavior of the synchronization errors satisfies additional criteria, such as requirements on a maximum settling time or a minimum damping. In this paper, it is first analyzed how these specifications on the the synchronization errors can be related to the parameters of the networked system. Secondly, a design method for synchronizing dynamic controllers is presented that ensures the requirements on the networked multi-agent system for arbitrary bidirectional interconnections. The results are demonstrated by its experimental application to the synchronization of the lateral dynamics of mobile robots.