Robustness and Consistency in Linear Quadratic Control with Predictions
Lists
Only you see this
Abstract
We study the problem of learning-augmented predictive linear quadratic control. Our goal is to design a controller that balances \textit{"consistency"}, which measures the competitive ratio when predictions are accurate, and \textit{"robustness"}, which bounds the competitive ratio when predictions are inaccurate. We propose a novel $\lambda$-confident policy and provide a competitive ratio upper bound that depends on a trust parameter $\lambda\in [0,1]$ set based on the confidence in the predictions and some prediction error $\varepsilon$. Motivated by online learning methods, we design a self-tuning policy that adaptively learns the trust parameter $\lambda$ with a competitive ratio that depends on $\varepsilon$ and the variation of system perturbations and predictions. We show that its competitive ratio is bounded from above by $ 1+{O(\varepsilon)}/({{\Theta(1)+\Theta(\varepsilon)}})+O(\mu_{\mathsf{Var}})$ where $\mu_\mathsf{Var}$ measures the variation of perturbations and predictions. It implies that when the variations of perturbations and predictions are small, by automatically adjusting the trust parameter online, the self-tuning scheme ensures a competitive ratio that does not scale up with the prediction error $\varepsilon$.
BibTeX
@article{li2022robustness,
author = {Tongxin Li and Ruixiao Yang and Guannan Qu and Guanya Shi and Chenkai Yu and Adam Wierman and Steven H. Low},
title = {{Robustness and Consistency in Linear Quadratic Control with Predictions}},
journal = {Proc. ACM Meas. Anal. Comput. Syst.},
year = {2022},
doi = {10.1145/3508038},
}