Event‐triggered Asynchronous Distributed Model Predictive Control with Variable Prediction Horizon for Nonlinear Systems
Pengbiao Wang,Xuemei Ren,Dongdong Zheng
DOI: https://doi.org/10.1002/rnc.6595
IF: 3.8973
2023-01-01
International Journal of Robust and Nonlinear Control
Abstract:In this article, we develop an event-triggered asynchronous distributed model predictive control (ETADMPC) algorithm with the adaptive prediction horizon for distributed nonlinear systems with weakly dynamics couplings, bounded disturbances, and system constraints. First, we focus on designing a novel adaptive event-triggered mechanism with Zeno-free phenomenons in order to reduce computational burdens, whose triggering threshold can adapt to the real-time changes of the system and make necessary adjustments. Then, a robust time-varying tightened state constraint is tailored for the optimization problem with respect to distributed model predictive control, and it can provide robustness to external disturbances and system coupling parts. And an adaptive prediction horizon update scheme is deliberately designed to decrease the length of the prediction horizon when the system state is close to the terminal set, reducing the computational complexity in the optimization problem. Furthermore, we strictly prove that under the given sufficient conditions, the proposed ETADMPC algorithm is recursively feasible and the closed-loop system is stable. Finally, a numerical example is provided to show that our scheme can achieve satisfactory control performances with less calculation and a shorter calculation time than the existing results.