Physics-informed Dyna-Style Model-Based Deep Reinforcement Learning for Dynamic Control

TL;DR

This paper introduces a physics-informed model-based reinforcement learning framework that leverages known physical laws to improve model accuracy and sample efficiency in control tasks governed by differential equations.

Contribution

It integrates physical knowledge into Dyna-style RL, enhancing model accuracy and reducing environment interactions for better control performance.

Findings

Improved sample efficiency in classic control problems.

Enhanced model accuracy using physics-informed constraints.

Demonstrated effectiveness on environments governed by differential equations.

Abstract

Model-based reinforcement learning (MBRL) is believed to have much higher sample efficiency compared to model-free algorithms by learning a predictive model of the environment. However, the performance of MBRL highly relies on the quality of the learned model, which is usually built in a black-box manner and may have poor predictive accuracy outside of the data distribution. The deficiencies of the learned model may prevent the policy from being fully optimized. Although some uncertainty analysis-based remedies have been proposed to alleviate this issue, model bias still poses a great challenge for MBRL. In this work, we propose to leverage the prior knowledge of underlying physics of the environment, where the governing laws are (partially) known. In particular, we developed a physics-informed MBRL framework, where governing equations and physical constraints are utilized to inform the model learning and policy search. By incorporating the prior information of the environment, the quality of the learned model can be notably improved, while the required interactions with the environment are significantly reduced, leading to better sample efficiency and learning performance. The effectiveness and merit have been demonstrated over a handful of classic control problems, where the environments are governed by canonical ordinary/partial differential equations.