Event‐triggered‐based secure adaptive NN consensus tracking control of switched multi‐agent systems

Abstract

AbstractThis paper addresses the problem of event‐triggered‐based secure adaptive neural‐networked (NN) consensus tracking control for a class of switched nonlinear multi‐agent systems (MASs) with unknown control directions under network attacks. In order to mitigate the effects of sensor and actuator attacks, a new event‐triggered‐based secure adaptive NN consensus tracking control protocol is proposed by exploring backstepping and the average dwell time method and the Nussbaum gain technique. In addition, if the switching happens between any two consecutive triggering instants, then asynchronous switching between subsystems of each follower and candidate controllers of subsystems will arise. Without any strict limitation on maximum asynchronous time, the asynchronous switching is handled by constructing a novel switching event‐triggered mechanism (ETM) and event‐triggered‐based adaptive NN controllers of subsystems for each follower. Meanwhile, the protocol proposed can guarantee that consensus tracking errors converge to a small neighborhood of the origin. Also, a strictly positive lower bound on inter‐event time is guaranteed. Finally, a practical example is presented to illustrate the effectiveness of the developed protocol.