Metraipolis › archive › dc_edbfe1768ee9

Real-World Reinforcement Learning with MPC Scaffolding for Dexterous Manipulation — arXiv:2609.14878

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.14878
License (as archived)
CC0-1.0
Archived
2026-10-16T02:14:41.039Z
SHA-256 of the archived text
f236a333286a1c79a23f18d8bfd68aaf66e53c6012b311e6618d3e725fb4c0fc
Stable URL of this page
https://metraipolis.com/documento/dc_edbfe1768ee9

How to cite

Real-World Reinforcement Learning with MPC Scaffolding for Dexterous Manipulation — arXiv:2609.14878 — arXiv (en), https://arxiv.org/abs/2609.14878 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_edbfe1768ee9 (Metraipolis archive dc_edbfe1768ee9, sha256 f236a333286a1c79a23f18d8bfd68aaf66e53c6012b311e6618d3e725fb4c0fc)
BibTeX
@misc{dcedbfe1768ee9,
  title = {Real-World Reinforcement Learning with MPC Scaffolding for Dexterous Manipulation — arXiv:2609.14878},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.14878},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_edbfe1768ee9; sha256 f236a333286a1c79a23f18d8bfd68aaf66e53c6012b311e6618d3e725fb4c0fc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14878. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Real-world reinforcement learning (RL) offers a promising route to dexterous manipulation policies that can adapt directly from physical interaction, but learning is hindered by inefficient early exploration and costly failures. We propose a framework that uses sampling-based model predictive control (MPC) as scaffolding for real-world dexterous RL, providing structured prior experience and task-directed guidance during learning without human demonstrations or corrective actions. A small set of MPC trajectories is first used to populate an offline replay buffer and to pretrain the actor and critic. During online learning, MPC intermittently guides data collection while an off-policy Soft Actor-Critic learner trains from both prior MPC experience and newly collected physical interaction, with control gradually transitioning to the learned policy. On continuous in-hand rotation with a 16-DoF Allegro hand, initialized from 20 MPC trajectories collected in 12 minutes on hardware, the policy reaches 100\% success after 7 minutes of online RL, with about three object drops on average during training. After 20 minutes of online learning, the policy achieves more than five times the rotation speed of the MPC controller and completes 1000 consecutive rotations without a drop. Ablations show complementary benefits from MPC-based pretraining, retained MPC experience, and online MPC guidance, while additional experiments demonstrate rapid adaptation to new object geometries and successful goal-conditioned reorientation.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.14878 — 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 2026-10-16T01:52:32.345Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).