An event-triggered model predictive control with exponentially stable offset free for PWA systems with model-plant mismatch. Issue 7 (May 2021)

• Record Type: Journal Article
• Authors: Li, Bingyun; Song, Chunyue; Zhao, Jun; Xu, Zuhua
• Is Part Of: Journal of the Franklin Institute. Volume 358:Issue 7(2021)
• Page Start: 3585
• Page End: 3608
• Publication Date: 2021-05
• Subjects: Science -- Periodicals; Technology -- Periodicals; Patents -- United States -- Periodicals; 505
• DOI: 10.1016/j.jfranklin.2021.02.036
• Languages: English
• ISSNs: 0016-0032

Abstract

Abstract: A new event-triggered model predictive control (MPC) method is proposed for PWA systems with model-plant mismatch, such that, for a given asymptotically constant reference signal, exponential stability and unbiased tracking can be achieved. Firstly, for the PWA system, an observer is designed to estimate state and the model-plant mismatch. To guarantee the exponential stability, an event is introduced, and an event-triggered model predictive controller is designed as well. By introducing state error and estimate error, an augmented model is constructed. Then, using the model-dependent average dwell time (MDADT) method and Lyapunov stability theory, exponentially stable condition of the closed-loop system is derived, which is formulated by linear matrix inequalities (LMIs), and zero offset is also guaranteed under some mild assumptions. Moreover, the MDADT of each sub-system is given simultaneously. Finally, simulations have been taken to verify the effectiveness of the proposed method.