Metraipolis › archive › dc_0200e16d9f61
RFPO: Rectified Flow Policy Optimization for Embodied Control — arXiv:2610.10453
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.10453
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T12:49:27.135Z
- SHA-256 of the archived text
dc2a1f681f2fef931958bc403a30de6d2e4502a2987c235e46675ff583c3dab2- Stable URL of this page
https://metraipolis.com/documento/dc_0200e16d9f61
How to cite
RFPO: Rectified Flow Policy Optimization for Embodied Control — arXiv:2610.10453 — arXiv (en), https://arxiv.org/abs/2610.10453 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0200e16d9f61 (Metraipolis archive dc_0200e16d9f61, sha256 dc2a1f681f2fef931958bc403a30de6d2e4502a2987c235e46675ff583c3dab2)
BibTeX
@misc{dc0200e16d9f61,
title = {RFPO: Rectified Flow Policy Optimization for Embodied Control — arXiv:2610.10453},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.10453},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0200e16d9f61; sha256 dc2a1f681f2fef931958bc403a30de6d2e4502a2987c235e46675ff583c3dab2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10453. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Flow-based policies provide an expressive framework for continuous robot control, but their iterative ODE integration incurs substantial inference cost. Naively reducing the integration budget can severely degrade control, since policies optimized under full-step execution are not explicitly constrained to remain reliable under coarse numerical integration. We refer to this mismatch as the few-step discretization gap. To address this problem, we introduce RFPO, a flow-policy optimization framework for reliable few-step execution. Reward-aware online Reflow rectifies student-induced transport paths during on-policy learning, making the resulting policy more robust to coarse integration. A frozen Gaussian PPO controller supplies complementary action-space supervision at full and intermediate integration budgets, while the deployed policy remains a single flow student executed with one Euler step. Across Unitree Go2, Boston Dynamics Spot, Unitree H1, and Unitree G1, RFPO consistently preserves full-step control performance under one-step execution, with one-step returns remaining within 2.4% of their corresponding 64-step values across both zero and random initialization. On Unitree Go2, one-step execution retains 98.5% of the 64-step reward while reducing onboard mean inference latency from 4.39 ms to 0.08 ms, yielding a 54.9x speedup. Real-robot experiments further validate stable one-step locomotion. Code: this https URL . Website: this https URL . Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2610.10453 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T12:18:30.586Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).