Fixed-relative-switch strategies for learning based event-triggered control of nonlinear multiagent systems

Ziming Wang,Yun gao,Apostolos I. Rikos,Xin Wang,Yiding Ji
2024-12-02
Abstract:This paper investigates event-triggered control for consensus tracking in nonlinear semi-strict-feedback multi-agent systems (MASs) with unknown states and subject to disturbances. We begin by employing radial basis function neural networks combined with the backstepping method to approximate the unknown nonlinear dynamics of the MASs, which facilitates the development of state and disturbance observers for estimation. We then propose three adaptive event-triggered control strategies: fixed-threshold, relative-threshold, and switch-threshold, whose controllers and triggering conditions are defined acocrdingly. By applying Lyapunov stability theory, we rigorously prove that all follower agents' outputs consistently track the reference of their leader, and all error signals remain uniformly bounded. Moreover, our control strategies effectively mitigate the occurrence of Zeno behaviors. To validate our control framework, an illustrative example is provided to demonstrate its desirable performance in consensus tracking and triggering efficiency.
Systems and Control
What problem does this paper attempt to address?