Control Analysis and Synthesis of Data-Driven Learning: A Kalman State-Space Approach

TL;DR

This paper introduces a Kalman state-space method for data-driven control that handles unknown plant models and uncertainties, enabling robust tracking through an ESO-based approach.

Contribution

It presents a novel Kalman state-space framework connecting data-driven and model-based control, incorporating ESO to manage uncertainties in iterative learning control.

Findings

Ensures model-free systems can track desired targets reliably.

Bridges data-driven control with classical model-based methods.

Validated through an example in iterative learning control.

Abstract

This paper aims to deal with the control analysis and synthesis problem of data-driven learning, regardless of unknown plant models and iteration-varying uncertainties. For the tracking of any desired target, a Kalman state-space approach is presented to transform it into two robust stability problems, which bridges a connection between data-driven control and model-based control. This approach also makes it possible to employ the extended state observer (ESO) in the design of data-driven learning to overcome the effect of iteration-varying uncertainties. It is shown that ESO-based data-driven learning ensures model-free systems to achieve the tracking of any desired target. In particular, our results apply to iterative learning control, which is verified by an example.