Output Feedback MPC for Nonlinear System in Large Operation Range

• 成果类型: Article
• 署名作者: Hu, Jianchen; Ding, Baocang
• 署名单位: Xi'an Jiaotong University; Chongqing University of Posts & Telecommunications
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3247878
• 发表日期: 2023
• 页码: 7903-7910
• 关键词: Model predictive control (MPC); Nonlinear model; Output feedback; recursive feasibility; STABILITY

摘要

This article studies the output feedback model predictive control (MPC) being suitable to a large operation range of the constrained system represented by a nonlinear model with a bounded disturbance. The model is further parameterized as a multiple linear parameter varying (LPV) state-space equation, each LPV representation being valid around an exclusive equilibrium. Offline, for each equilibrium, a set of local control laws (in the dynamic output feedback), each having its region of attraction, are calculated. All the control laws for all the equilibriums constitute a lookup table, and all the corresponding regions of attraction take a union. Online, the real-time control law is searched as one, as per its region of attraction, in the lookup table when the estimated system state moves inside of the union. Once the initial estimation error lies in a prespecified set, the augmented state is guaranteed to converge to the neighborhood of the target equilibrium, and the constraints on input and output are consistently satisfied. An example is given to illustrate the effectiveness of the proposed algorithm.