Metraipolis › archive › dc_224e55f995dc
ProxPI: Proximal Prior Injection for Sampling-Based MPC under Learned-Prior Mismatch — arXiv:2609.00941
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.00941
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:18:25.549Z
- SHA-256 of the archived text
f1820ee0c65e480544d5612b407c59b129557565725e52c93f6e71218e85fbfe- Stable URL of this page
https://metraipolis.com/documento/dc_224e55f995dc
How to cite
ProxPI: Proximal Prior Injection for Sampling-Based MPC under Learned-Prior Mismatch — arXiv:2609.00941 — arXiv (en), https://arxiv.org/abs/2609.00941 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_224e55f995dc (Metraipolis archive dc_224e55f995dc, sha256 f1820ee0c65e480544d5612b407c59b129557565725e52c93f6e71218e85fbfe)
BibTeX
@misc{dc224e55f995dc,
title = {ProxPI: Proximal Prior Injection for Sampling-Based MPC under Learned-Prior Mismatch — arXiv:2609.00941},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.00941},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_224e55f995dc; sha256 f1820ee0c65e480544d5612b407c59b129557565725e52c93f6e71218e85fbfe},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00941. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Combining learned policies with model predictive control can leverage learned task priors while retaining online adaptation to new objectives and constraints, but performance degrades when the policy is out of distribution. In policy-guided model predictive path integral (MPPI) control, a policy-centered warm-start approach centers the sampling distribution on the policy output. When the prior is mismatched, centering the sampling distribution on the policy output restricts exploration around an unsuitable solution and prevents recovery toward the task optimum. We propose Proximal Prior Injection (ProxPI), which retains nominal-centered MPPI sampling and incorporates the policy through a soft proximity cost. This matches the in-distribution performance of existing prior-injection schemes while enabling the optimizer to escape an inaccurate policy and recover vanilla MPPI-level performance. We theoretically show that re-centering on the prior discards the optimizer's correction at every update, whereas nominal-centered sampling retains it and converges to a solution set by both the task cost and the prior, and that this failure is not removed by a larger rollout budget. Simulations and real-robot experiments demonstrate robust performance under both in-distribution and out-of-distribution tasks. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.RO). Categorie arXiv: Robotics (cs.RO) ; Systems and Control (eess.SY). Pagina del lavoro: https://arxiv.org/abs/2609.00941 — 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 2026-10-17T07:06:32.694Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).