Event-triggered ISS-modular neural network control for containment maneuvering of nonlinear strict-feedback multi-agent systems

Abstract

The paper is concerned with containment maneuvering of nonlinear multi-input multi-output (MIMO) strict-feedback multi-agent systems under the distributed directed communication. Followers are driven into a convex hull spanned by all virtual leaders guided by desired parameterized paths. Multi-agent systems are driven to achieve the desired behavior. A containment maneuvering controller is developed based on an event-triggered modular-ISS neural network control method to decouple estimation and control. Specifically, in the estimation loop, an RBF-network-based predictor module is constructed to design the adaptation law, where an RBF network is employed to identify the uncertain nonlinear dynamics. An event-triggered communication mechanism is introduced to decrease the burden of followers communication. In the control loop, a control module is constructed by using the output of predictor module and a third-order linear tracking differentiator. An event-triggered actuation mechanism is introduced to reduce the actuator burden of followers. A path update law is constructed to achieve coordination among leaders. An event-triggered communication scheme is introduced for leaders to decrease the communication burden of leaders. The input-to-state stability of the total cascade system is demonstrated by using the cascade stability theorem. Simulation results are provided to substantiate the efficacy of the proposed containment maneuvering controller for nonlinear MIMO strict-feedback multi-agent systems.