Distributed optimized dynamic event-triggered control for unknown heterogeneous nonlinear MASs with input-constrained

Lina Xia,Qing Li,Ruizhuo Song,Shuzhi Sam Ge
DOI: https://doi.org/10.1016/j.neunet.2022.06.033
Abstract:The distributed optimized dynamic event-triggered controller is investigated for completely unknown heterogeneous nonlinear multi-agent systems (MASs) on a directed graph subject to input-constrained. First, the distributed observer is designed to estimate the information of the leader for each follower, and a network of the augmented system is constructed by employing the dynamics of the followers and the observers. An identifier with a compensator is designed to approximate the unknown augmented system (agent) with an arbitrarily small identifier error. Then, consider that the input-constrained optimal controller, along with Hamilton-Jacobi-Bellman (HJB) equation, is under pressure to execute in certain systems associated with bottlenecks such as communication and computing burdens. A critic-actor-based optimized dynamic event-triggered controller, which tunes the parameters of critic-actor neural networks (NNs) by the dynamic triggering mechanism, is leveraged to determine the rule of aperiodic sampling and maintain the desired synchronization service. In addition, the existence of a positive minimum inter-event time (MIET) between consecutive events is also proved. Finally, the applications in non-identical nonlinear MAS and 2-DOF robots illustrate the availability of the proposed theoretical results.
What problem does this paper attempt to address?