Stochastic Data-Driven Predictive Control: Chance-Constraint Satisfaction With Identified Multi-Step Predictors

作     者：Balim, Haldun Carron, Andrea Zeilinger, Melanie N. Kohler, Johannes 

作者机构：Swiss Fed Inst Technol Inst Dynam Syst & Control CH-8052 Zurich Switzerland Sch Engn & Appl Sci Harvard MA 02138 USA 

出 版 物：《IEEE CONTROL SYSTEMS LETTERS》 (IEEE Control Syst. Lett.)

年 卷 期：2024年第8卷

页      面：3249-3254页

核心收录：

基　　金：Swiss National Science Foundation [51NF40 180545] ETH Career Seed Award through the ETH Zurich Foundation 

主　　题：Uncertainty Kalman filters State-space methods Stochastic processes Technological innovation Predictive models Predictive control Maximum likelihood estimation Noise Noise measurement Predictive control for linear systems data driven control identification for control 

摘      要：We propose a novel data-driven stochastic model predictive control framework for uncertain linear systems with noisy output measurements. Our approach leverages multi-step predictors to efficiently propagate uncertainty, ensuring chance constraint satisfaction. In particular, we present a strategy to identify multi-step predictors and quantify the associated uncertainty using a surrogate (data-driven) state space model. Then, we utilize the derived distribution to formulate a constraint tightening that ensures chance constraint satisfaction despite the parametric uncertainty. A numerical example highlights the reduced conservatism of handling parametric uncertainty in the proposed method compared to state-of-the-art solutions.