Distributionally Robust Data-Driven Predictive Control for Stochastic LTI Systems

TL;DR

This paper introduces a distributionally robust predictive control method for stochastic LTI systems that uses data-driven distribution estimates and Wasserstein ambiguity sets to ensure probabilistic constraints and cost bounds.

Contribution

It develops a tractable, data-driven control framework that avoids explicit predictor identification by leveraging Wasserstein ambiguity sets and finite-sample concentration results.

Findings

The proposed method provides probabilistic guarantees on cost and constraints.

Numerical simulations show improved robustness over existing methods.

The approach is applicable under various disturbance conditions and cost functions.

Abstract

We propose a distributionally robust data-driven predictive control framework for stochastic linear time-invariant systems with unknown dynamics and disturbance distributions. We use an offline trajectory to fit the subspace predictive control (SPC) predictor via least squares and construct an empirical distribution of the prediction residuals as a proxy for the unknown disturbance distribution. We then center a Wasserstein ambiguity set around this estimate and minimize the worst-case expected cost while enforcing probabilistic output constraint satisfaction over all distributions in the set. The resulting problem admits a tractable reformulation with an equivalent direct data-driven form, eliminating the need for explicit predictor identification. Using finite-sample concentration results, we provide a data-driven Wasserstein radius such that, with high probability, the true expected cost is bounded above by the tractable objective and output constraints are satisfied with respect to the true disturbance distribution. Numerical simulations validate the framework against existing methods under various disturbance conditions and cost functions.