Abstract:This paper investigates the issue of event-triggered distributed model predictive control for coupled nonlinear systems with additive disturbances. Specifically, this paper proposes two event-triggered strategies, which are incorporated into the sensor and model predictive control (MPC) based controller for each subsystem, respectively. A limited-information-based control scheme is constructed using two-channel even-triggered transmissions. The scheme proposed achieves efficient reduction in both the transmission rates of the sensor and the resource consumption associated with optimization problem, as well as, enhances the real-world operational capability through the utilization of a sample-and-hold technique. This technique allows the actual control inputs to be derived by discretizing the continuous optimal control trajectory. This paper shows rigorously that the mutual influences invoked by dynamic coupling are bounded and the Zeno behavior is excluded entirely. Also, the sufficient conditions are developed to ensure the algorithm feasibility and the convergence of the overall system to a bounded set. Finally, a practical example is presented and comparisons are made to demonstrate the efficiency of the proposed algorithm.

Distributed Model Predictive Control for Coupled Nonlinear Systems via Two-Channel Event- Triggered Transmission Scheme