Distributed Model Predictive Control for Nonlinear Networked Systems with Asynchronous Communication

摘要

This paper concerns the distributed model predictive control (DMPC) problem for nonlinear multirate networked systems subject to asynchronous communication. The asynchronization of communication between each subsystem always exists in mechanic engineering and chemical processes. The motivation of this paper is to use the asynchronous information to improve closed-loop performance for a class of networked systems with coupled nonlinear dynamics. To handle the asynchronous communication, a methodology of state trajectory estimation for the coupled neighboring agents over each prediction horizon with guaranteed error bounds is proposed. To make effective use of computational resources, a distributed model predictive control algorithm with the dual-model control strategy is designed. The conditions for ensuring the subsequent feasibility of the proposed algorithm and the stabilization of the closed-loop networked systems are developed respectively. Finally, the theoretical results are verified by a simulation example for a set of coupled Van der Pol oscillators.