Embracing Flow Unsteadiness: A High-Throughput Learning Platform Enables Vortex-Exploiting Bioinspired Propulsion
Abstract
Biological swimmers and flyers exploit unsteady vortices for propulsion, whereas engineered vehicles usually suppress them as disturbances. Learning such flow exploitation in machines is difficult because real-fluid interaction data are scarce and unstructured exploration is unstable in high-dimensional, history-dependent flows. Here we present REEF, a co-designed physical-learning framework that integrates SHOAL, an eight-channel high-throughput array for real fluid--structure interaction, with V-STAR, a staged algorithm that converts these interactions into policies through imitation, offline internalization, and online adaptation. Across lift-based, drag-based, and momentum-jet propulsors, REEF expands the attainable force envelope to more than twice that of parameterized search. Particle image velocimetry shows that these gains arise from coordinated vortex formation, growth, and force projection, rather than refinement of a fixed motion-to-force mapping. Force-trained policies transfer zero-shot to free-moving robots whose body motion changes the surrounding flow, suggesting that REEF learns transferable wake-coupling principles for embodied propulsion in unsteady fluids.
Related
- Influence of Wind Unsteadiness and Inhomogeneity on the Mean and Fluctuating Aerodynamic Force Acting on a Wingsail
- DSMC Study of Hypersonic Boundary-Layer Unsteadiness Using Data-Driven Analysis
- DSMC Study on Unsteadiness of Supersonic Flow Over a Cylinder
- On flow unsteadiness in strongly separated high-speed ramp flows using kinetic and data-driven methods
- Optimizing twisted partially coherent vortex beams for ground-based laser ablation propulsion in space
BibTeX
@misc{han2026embracing,
title = {{Embracing Flow Unsteadiness: A High-Throughput Learning Platform Enables Vortex-Exploiting Bioinspired Propulsion}},
author = {Fei Han and Xinyu Cui and Zhipeng Wang and Ning Yang and Hang Xu and Haifeng Zhang and Zhongming Hu and Jun Wang and Junfeng Du and Dixia Fan},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.14254},
eprint = {2609.14254},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.14254},
}
RIS
TY - UNPB
TI - Embracing Flow Unsteadiness: A High-Throughput Learning Platform Enables Vortex-Exploiting Bioinspired Propulsion
AU - Fei Han
AU - Xinyu Cui
AU - Zhipeng Wang
AU - Ning Yang
AU - Hang Xu
AU - Haifeng Zhang
AU - Zhongming Hu
AU - Jun Wang
AU - Junfeng Du
AU - Dixia Fan
PY - 2026
DA - 2026/09/13
PB - arXiv
DO - 10.48550/arxiv.2609.14254
UR - https://arxiv.org/abs/2609.14254
AB - Biological swimmers and flyers exploit unsteady vortices for propulsion, whereas engineered vehicles usually suppress them as disturbances. Learning such flow exploitation in machines is difficult because real-fluid interaction data are scarce and unstructured exploration is unstable in high-dimensional, history-dependent flows. Here we present REEF, a co-designed physical-learning framework that integrates SHOAL, an eight-channel high-throughput array for real fluid--structure interaction, with V-STAR, a staged algorithm that converts these interactions into policies through imitation, offline internalization, and online adaptation. Across lift-based, drag-based, and momentum-jet propulsors, REEF expands the attainable force envelope to more than twice that of parameterized search. Particle image velocimetry shows that these gains arise from coordinated vortex formation, growth, and force projection, rather than refinement of a fixed motion-to-force mapping. Force-trained policies transfer zero-shot to free-moving robots whose body motion changes the surrounding flow, suggesting that REEF learns transferable wake-coupling principles for embodied propulsion in unsteady fluids.
ER -