Asynchronous Parallel Reinforcement Learning for Optimizing Propulsive Performance in Fin Ray Control

TL;DR

This paper presents an asynchronous parallel reinforcement learning method for optimizing fin-ray control in fish-like propulsion, demonstrating improved performance and training efficiency over traditional approaches.

Contribution

Introduces an innovative asynchronous parallel training strategy for DRL in fluid-structure interaction environments, enabling scalable and efficient fin-ray control policy learning.

Findings

Achieved superior propulsive performance compared to sinusoidal actuation.

Demonstrated the effectiveness of APT in training efficiency and scalability.

Validated the method through numerical experiments on nonlinear dynamics.

Abstract

Fish fin rays constitute a sophisticated control system for ray-finned fish, facilitating versatile locomotion within complex fluid environments. Despite extensive research on the kinematics and hydrodynamics of fish locomotion, the intricate control strategies in fin-ray actuation remain largely unexplored. While deep reinforcement learning (DRL) has demonstrated potential in managing complex nonlinear dynamics; its trial-and-error nature limits its application to problems involving computationally demanding environmental interactions. This study introduces a cutting-edge off-policy DRL algorithm, interacting with a fluid-structure interaction (FSI) environment to acquire intricate fin-ray control strategies tailored for various propulsive performance objectives. To enhance training efficiency and enable scalable parallelism, an innovative asynchronous parallel training (APT) strategy is proposed, which fully decouples FSI environment interactions and policy/value network optimization. The results demonstrated the success of the proposed method in discovering optimal complex policies for fin-ray actuation control, resulting in a superior propulsive performance compared to the optimal sinusoidal actuation function identified through a parametric grid search. The merit and effectiveness of the APT approach are also showcased through comprehensive comparison with conventional DRL training strategies in numerical experiments of controlling nonlinear dynamics.