How Training Data Impacts Performance in Learning-based Control

TL;DR

This paper establishes an analytical relationship between training data density and control performance in learning-based control, introducing a new data quality measure called the rho-gap and demonstrating its application.

Contribution

It introduces the rho-gap measure for training data quality and derives bounds on control performance considering data distribution and model uncertainty.

Findings

The rho-gap correlates data density with control accuracy.

Analytical bounds on tracking error are derived based on data quality.

Numerical examples validate the proposed measure and bounds.

Abstract

When first principle models cannot be derived due to the complexity of the real system, data-driven methods allow us to build models from system observations. As these models are employed in learning-based control, the quality of the data plays a crucial role for the performance of the resulting control law. Nevertheless, there hardly exist measures for assessing training data sets, and the impact of the distribution of the data on the closed-loop system properties is largely unknown. This paper derives - based on Gaussian process models - an analytical relationship between the density of the training data and the control performance. We formulate a quality measure for the data set, which we refer to as $\rho$-gap, and derive the ultimate bound for the tracking error under consideration of the model uncertainty. We show how the $\rho$-gap can be applied to a feedback linearizing control law and provide numerical illustrations for our approach.