Distributed adaptive control of linear multi-agent systems with event-triggered communications

This paper addresses the average consensus problem of multi-agent systems with general linear dynamics via event-triggered communications. It is shown how the design of both the control law and the trigger functions depends on parameters of the overall system, and an algorithm is proposed to estimate these parameters in a distributed fashion that does not imply an increase in the number of the communications between agents. As a result, it yields an adaptive control law and an update of the trigger functions' parameters only at event times. Proofs of asymptotic convergence to average consensus and existence of positive lower bound for the inter-event intervals are provided. Numerical simulations show the effectiveness of the proposed approach and how it compares to constant values of the parameters.