Boosting for Control of Dynamical Systems

TL;DR

This paper introduces a boosting framework for online control of dynamical systems, combining weak controllers into a stronger one with proven performance improvements, supported by empirical results.

Contribution

It presents an efficient boosting algorithm for aggregating controllers, advancing control performance through a novel online boosting approach.

Findings

The boosting algorithm improves control accuracy in various settings.

Theoretical guarantees for the combined controller's performance.

Empirical results validate the effectiveness of the proposed method.

Abstract

We study the question of how to aggregate controllers for dynamical systems in order to improve their performance. To this end, we propose a framework of boosting for online control. Our main result is an efficient boosting algorithm that combines weak controllers into a provably more accurate one. Empirical evaluation on a host of control settings supports our theoretical findings.