Data-driven Estimator Synthesis with Instantaneous Noise

TL;DR

This work directly deriving a suitable parameterization in primal space allows the integration of additional structural knowledge, such as bounds on parameters, in data-driven controller design based on the data-informativity framework.

Abstract

Data-driven controller design based on data informativity has gained popularity due to its straightforward applicability, while providing rigorous guarantees. However, applying this framework to the estimator synthesis problem introduces technical challenges, which can only be solved so far by adding restrictive assumptions. In this work, we remove these restrictions to improve performance guarantees. Moreover, our parameterization allows the integration of additional structural knowledge, such as bounds on parameters. Our findings are validated using numerical examples.

Figures (1)

• Figure 1: Visualization of $\Sigma$, $\Sigma_{0}$ and $\hat{\Sigma}$.

Theorems & Definitions (4)

• proof
• proof
• proof
• proof