Plug-and-Play Benchmarking of Reinforcement Learning Algorithms for Large-Scale Flow Control

TL;DR

FluidGym is a fully differentiable, GPU-accelerated benchmark suite for reinforcement learning in active flow control, enabling standardized, scalable, and accessible evaluation without external CFD software.

Contribution

This paper introduces FluidGym, the first standalone, fully differentiable RL benchmark for AFC built in PyTorch, addressing limitations of existing benchmarks.

Findings

Baseline results with PPO and SAC demonstrate effectiveness.

FluidGym enables systematic comparison of control methods.

All resources are publicly available for research use.

Abstract

Reinforcement learning (RL) has shown promising results in active flow control (AFC), yet progress in the field remains difficult to assess as existing studies rely on heterogeneous observation and actuation schemes, numerical setups, and evaluation protocols. Current AFC benchmarks attempt to address these issues but heavily rely on external computational fluid dynamics (CFD) solvers, are not fully differentiable, and provide limited 3D and multi-agent support. To overcome these limitations, we introduce FluidGym, the first standalone, fully differentiable benchmark suite for RL in AFC. Built entirely in PyTorch on top of the GPU-accelerated PICT solver, FluidGym runs in a single Python stack, requires no external CFD software, and provides standardized evaluation protocols. We present baseline results with PPO and SAC and release all environments, datasets, and trained models as public resources. FluidGym enables systematic comparison of control methods, establishes a scalable foundation for future research in learning-based flow control, and is available at https://github.com/safe-autonomous-systems/fluidgym.