Dynamic Event-Triggered Cooperative Adaptive Optimal Output Regulation for Multiagent Systems With Input Saturation
针对输入饱和的未知离散时间多智能体系统，提出一种动态事件触发协同自适应最优输出调节方法，通过分布式观测器和内模原理降低通信与计算成本，并保证控制输入在限制内。
This article investigates event-triggered cooperative adaptive optimal output regulation for unknown discrete-time multiagent systems (MASs) with input saturation. To address the issue that some followers may have no direct access to the leader, distributed observers are proposed to estimate the reference signals. A dynamic event-triggering mechanism is introduced to reduce communication and computational costs. By combining the internal model principle with low-gain and policy iteration (PI) techniques, an inner-outer loop-based dynamic event-triggered adaptive optimal control approach is developed. The convergence of the proposed algorithm is rigorously analyzed, and the control inputs are explicitly constrained within the input limits. A comprehensive stability analysis is provided, along with conditions for the MASs to achieve leader-to-formation stability (LFS). The sensitivity of the suboptimality index to system parameters is also taken into consideration. Finally, the effectiveness of the proposed approach is validated through a simulation example applied to grid-connected ac microgrid control.