Context-aware controller inference for stabilizing dynamical systems from scarce data

TL;DR

This paper presents a data-efficient, context-aware control method that stabilizes high-dimensional dynamical systems by focusing only on unstable dynamics, requiring significantly less data than traditional approaches.

Contribution

It introduces a novel controller inference technique that learns stabilizing controllers from scarce data by targeting unstable dynamics in high-dimensional systems.

Findings

Learns stabilizing controllers with orders of magnitude fewer data.

Effective in data-scarce, complex physics engineering problems.

Outperforms traditional data-driven control and reinforcement learning methods.

Abstract

This work introduces a data-driven control approach for stabilizing high-dimensional dynamical systems from scarce data. The proposed context-aware controller inference approach is based on the observation that controllers need to act locally only on the unstable dynamics to stabilize systems. This means it is sufficient to learn the unstable dynamics alone, which are typically confined to much lower dimensional spaces than the high-dimensional state spaces of all system dynamics and thus few data samples are sufficient to identify them. Numerical experiments demonstrate that context-aware controller inference learns stabilizing controllers from orders of magnitude fewer data samples than traditional data-driven control techniques and variants of reinforcement learning. The experiments further show that the low data requirements of context-aware controller inference are especially beneficial in data-scarce engineering problems with complex physics, for which learning complete system dynamics is often intractable in terms of data and training costs.