A critical assessment of reinforcement learning methods for microswimmer navigation in complex flows
arXiv:2505.05525 · doi:10.1140/epje/s10189-025-00522-2
Abstract
Navigating in a fluid flow while being carried by it, using only information accessible from on-board sensors, is a problem commonly faced by small planktonic organisms. It is also directly relevant to autonomous robots deployed in the oceans. In the last ten years, the fluid mechanics community has widely adopted reinforcement learning, often in the form of its simplest implementations, to address this challenge. But it is unclear how good are the strategies learned by these algorithms. In this paper, we perform a quantitative assessment of reinforcement learning methods applied to navigation in partially observable flows. We first introduce a well-posed problem of directional navigation for which a quasi-optimal policy is known analytically. We then report on the poor performance and robustness of commonly used algorithms (Q-Learning, Advantage Actor Critic) in flows regularly encountered in the literature: Taylor-Green vortices, Arnold-Beltrami-Childress flow, and two-dimensional turbulence. We show that they are vastly surpassed by PPO (Proximal Policy Optimization), a more advanced algorithm that has established dominance across a wide range of benchmarks in the reinforcement learning community. In particular, our custom implementation of PPO matches the theoretical quasi-optimal performance in turbulent flow and does so in a robust manner. Reaching this result required the use of several additional techniques, such as vectorized environments and generalized advantage estimation, as well as hyperparameter optimization. This study demonstrates the importance of algorithm selection, implementation details, and fine-tuning for discovering truly smart autonomous navigation strategies in complex flows.
References in corpus (16)
- Adaptive locomotion of artificial microswimmers
- Flow Navigation by Smart Microswimmers via Reinforcement Learning
- Learning to school in the presence of hydrodynamic interactions
- Zermelo's problem: Optimal point-to-point navigation in 2D turbulent flows using Reinforcement Learning
- Machine learning strategies for path-planning microswimmers in turbulent flows
- Finding Efficient Swimming Strategies in a Three Dimensional Chaotic Flow by Reinforcement Learning
- Searching for a source without gradients: how good is infotaxis and how to beat it
- Smart Inertial Particles
- Navigation of micro-swimmers in steady flow: the importance of symmetries
- Surfing on turbulence: a strategy for planktonic navigation
- Deep reinforcement learning for the olfactory search POMDP: a quantitative benchmark
- Optimal tracking strategies in a turbulent flow
- Energetic cost of microswimmer navigation: the role of body shape
- Taming Lagrangian Chaos with Multi-Objective Reinforcement Learning
- Learning to swim efficiently in a nonuniform flow field
- Short term vs. long term: optimization of microswimmer navigation on different time horizons