Model predictive control for complicated dynamic systems: a survey
Yan Song, Bin Zhang, Chuanbo Wen, Dong Wang, Guoliang Wei
Abstract
Nowadays, model predictive control (MPC) has emerged as a powerful technique for controlling complex and dynamic systems, exhibiting advantages over traditional control strategies in terms of performance and robustness. This survey aims to provide a comprehensive overview of MPC theory and its applications in a variety of challenging dynamic systems. Firstly, the relationship between MPC and traditional optimal control is revealed by interpreting the basic research concepts of the qualitative synthesis theory of MPC. Then, starting from the basic problem of MPC without model uncertainty, a series of MPC methods and basic research ideas for solving systems with model uncertainty are presented. Furthermore, three classes of typical complicated control systems (networked systems, Markov jump systems and Takagi-Sugeno fuzzy systems) are introduced and the key issues in adopting MPC methods for these classes of systems are illustrated. Finally, a summary and future topics are provided.