Event-based Consensus Control of Multi-agent Systems by L ∞ Theory

This paper investigates the consensus problem of multi-agent systems with general linear models by the event-based control strategy, and develops the protocol design method using the L (a) control theory. A distributed event-based protocol, which yields the control inputs using the estimated states based on local information at event times, is presented for the asynchronously event-triggered communication network. Here, the state estimation errors, arising from the event-based interconnection, are regarded as external disturbances with limited peak values. Then by defining a controlled output that measures the state disagreement among agents, the consensus condition and controller design criterion are obtained by the L (a) theory of disturbed systems. It shows that all agents will converge to a desired bounded region around the consensus trajectory under the proposed protocol.