Saved in:
Bibliographic Details
Main Authors: Mattsson, Per, Bonassi, Fabio, Breschi, Valentina, Schön, Thomas B.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.05860
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929347713040384
author Mattsson, Per
Bonassi, Fabio
Breschi, Valentina
Schön, Thomas B.
author_facet Mattsson, Per
Bonassi, Fabio
Breschi, Valentina
Schön, Thomas B.
contents Recently, several direct Data-Driven Predictive Control (DDPC) methods have been proposed, advocating the possibility of designing predictive controllers from historical input-output trajectories without the need to identify a model. In this work, we show that these approaches are equivalent to an indirect approach. Reformulating the direct methods in terms of estimated parameters and covariance matrices allows us to give new insights into how they work in comparison with, for example, Subspace Predictive Control (SPC). In particular, we show that for unconstrained problems the direct methods are equivalent to SPC with a reduced weight on the tracking cost. Via a numerical experiment, motivated by the reformulation, we also illustrate why the performance of direct DDPC methods with fixed regularization tends to degrade as the number of training samples increases.
format Preprint
id arxiv_https___arxiv_org_abs_2403_05860
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the equivalence of direct and indirect data-driven predictive control approaches
Mattsson, Per
Bonassi, Fabio
Breschi, Valentina
Schön, Thomas B.
Systems and Control
Recently, several direct Data-Driven Predictive Control (DDPC) methods have been proposed, advocating the possibility of designing predictive controllers from historical input-output trajectories without the need to identify a model. In this work, we show that these approaches are equivalent to an indirect approach. Reformulating the direct methods in terms of estimated parameters and covariance matrices allows us to give new insights into how they work in comparison with, for example, Subspace Predictive Control (SPC). In particular, we show that for unconstrained problems the direct methods are equivalent to SPC with a reduced weight on the tracking cost. Via a numerical experiment, motivated by the reformulation, we also illustrate why the performance of direct DDPC methods with fixed regularization tends to degrade as the number of training samples increases.
title On the equivalence of direct and indirect data-driven predictive control approaches
topic Systems and Control
url https://arxiv.org/abs/2403.05860