Data-driven model predictive control: closed-loop guarantees and experimental results
TL;DR: In this article, the authors provide a comprehensive review and practical implementation of a recently developed model predictive control (MPC) framework for controlling unknown systems using only measured data and no explicit model knowledge, relying on an implicit system parametrization from behavioral systems theory based on one measured input-output trajectory.
read more
Abstract: We provide a comprehensive review and practical implementation of a recently developed model predictive control (MPC) framework for controlling unknown systems using only measured data and no explicit model knowledge. Our approach relies on an implicit system parametrization from behavioral systems theory based on one measured input-output trajectory. The presented MPC schemes guarantee closed-loop stability for unknown linear time-invariant (LTI) systems, even if the data are affected by noise. Further, we extend this MPC framework to control unknown nonlinear systems by continuously updating the data-driven system representation using new measurements. The simple and intuitive applicability of our approach is demonstrated with a nonlinear four-tank system in simulation and in an experiment.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Behavioral systems theory in data-driven analysis, signal processing, and control
Ivan Markovsky,Florian Dörfler +1 more
TL;DR: Data-driven analysis, signal processing, and control methods as mentioned in this paper can be broadly classified as implicit and explicit approaches, with the implicit approach being more robust to uncertainty and robustness to noise.
173
•Posted Content
Linear tracking MPC for nonlinear systems Part II: The data-driven case.
TL;DR: In this article, a data-driven MPC approach to control unknown nonlinear systems using only measured input-output data with closed-loop stability guarantees is presented. But this approach is limited to affine systems.
18
•Posted Content
Adaptive Robust Data-driven Building Control via Bi-level Reformulation: an Experimental Result.
Yingzhao Lian,Jicheng Shi,Manuel Pascal Koch,Colin N. Jones +3 more
- 10 Jun 2021
TL;DR: In this article, a robust bilevel formulation for data-driven adaptive building control with measured process noise and unknown measurement noise via a robust BLEW formulation is proposed. But the authors do not consider the effect of noise on the prediction quality.
16
•Posted Content
On a Stochastic Fundamental Lemma and Its Use for Data-Driven MPC
Guanru Pan,Ruchuan Ou,Timm Faulwasser +2 more
- 26 Nov 2021
TL;DR: In this paper, the authors propose a novel variant of the fundamental lemma for stochastic LTI systems, which allows to predict future distributions of the behaviors of LTI system based on the knowledge of previously recorded behavior realizations and based on knowledge of the noise distribution.
References
A note on persistency of excitation
TL;DR: It is proved that if a component of the response signal of a controllable linear time-invariant system is persistently exciting of sufficiently high order, then the windows of the signal span the full system behavior.
1K
Learning-Based Model Predictive Control: Toward Safe Learning in Control
TL;DR: This research presents a meta-modelling framework that automates the very labor-intensive and therefore time-heavy and expensive process of manually cataloging and cataloging individual neurons to provide real-time information about their levels of activity.
758
Data-Driven Model Predictive Control With Stability and Robustness Guarantees
TL;DR: The presented results provide the first (theoretical) analysis of closed-loop properties, resulting from a simple, purely data-driven MPC scheme, including a slack variable with regularization in the cost.
Data-Enabled Predictive Control: In the Shallows of the DeePC
Jeremy Coulson,John Lygeros,Florian Dörfler +2 more
- 25 Jun 2019
TL;DR: In this paper, a data-enabled predictive control (DeePC) algorithm is presented that computes optimal and safe control policies using real-time feedback driving the unknown system along a desired trajectory while satisfying system constraints.
Provably safe and robust learning-based model predictive control
TL;DR: A learning-based model predictive control scheme that provides deterministic guarantees on robustness, while statistical identification tools are used to identify richer models of the system in order to improve performance.
608