Event‐triggered distributed model predictive control for PWA systems

Abstract The online computational burden and control performance are two main issues in implementing distributed model predictive control for piecewise affine systems. In the previous research, many methods have been proposed to solve these issues, such as a one‐step distributed model predictive control method that was proposed to reduce the heavy online computational burden. However, its control performance is limited because of the one‐step prediction horizon. This article proposes an event‐triggered distributed model predictive control algorithm for such systems. The online computational burden is alleviated within the distributed framework through the use of an event‐triggered mechanism and a variable prediction horizon approach. These strategies not only reduce the number of optimization problems requiring online resolution and simplify them, but also allow for a balance between the control performance and the online computational burden by adjusting event‐triggering thresholds. The algorithm utilizes state tubes and terminal sets that are tailored to the characteristics of the piecewise affine systems, thereby rigorously establishing the recursive feasibility of the optimization problems and the stability of the closed‐loop system. The simulation results validate the effectiveness of the proposed algorithm.