Distributed MPC with composite event-triggering mechanisms for nonlinear multi-agent systems

In this paper, an event-triggered distributed model predictive control (ET-DMPC) strategy is investigated for nonlinear multi-agent systems (MASs) with additive disturbances. Firstly, based on the information of system state errors, composite event-triggering mechanisms (ETMs) are designed to achieve a significant decrease in the information transmission rate at the sensor nodes and the computational and communication burden at the controller nodes. Secondly, a reconstruction mechanism based on zero-order hold fashion is established for continuous control trajectories to effectively reduce the bandwidth occupation during network channel transmission of control signals. Then, through rigorous theoretical analysis, sufficient conditions are obtained to guarantee the feasibility of algorithms, the closed-loop stability of considered systems, and the Zeno-free behavior. In the end, the effectiveness of the proposed algorithms is verified through simulation and comparison.