On the impact of regularization in data-driven predictive control

TL;DR

This paper investigates how different regularization penalties affect the performance of data-driven predictive control methods, providing guidelines for tuning in various data and noise conditions.

Contribution

It analyzes the impact of regularization choices on $oldsymbol{ extgamma}$-DDPC's performance and offers practical tuning guidelines for different scenarios.

Findings

Regularization significantly influences control performance.

Guidelines for tuning regularization coefficients are provided.

Performance varies with data quality and noise levels.

Abstract

Model predictive control (MPC) is a control strategy widely used in industrial applications. However, its implementation typically requires a mathematical model of the system being controlled, which can be a time-consuming and expensive task. Data-driven predictive control (DDPC) methods offer an alternative approach that does not require an explicit mathematical model, but instead optimize the control policy directly from data. In this paper, we study the impact of two different regularization penalties on the closed-loop performance of a recently introduced data-driven method called $\gamma$-DDPC. Moreover, we discuss the tuning of the related coefficients in different data and noise scenarios, to provide some guidelines for the end user.