Metraipolis › archive › dc_abe7b6dd33c4

Learning In-Hand Object Reaching to General 6D Poses — arXiv:2609.13761

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13761
License (as archived)
CC0-1.0
Archived
2026-10-16T02:09:26.050Z
SHA-256 of the archived text
845631b5f898447de9af35bb905e5c61ec1bc20f2581f7e71ee4fc8751b129a0
Stable URL of this page
https://metraipolis.com/documento/dc_abe7b6dd33c4

How to cite

Learning In-Hand Object Reaching to General 6D Poses — arXiv:2609.13761 — arXiv (en), https://arxiv.org/abs/2609.13761 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_abe7b6dd33c4 (Metraipolis archive dc_abe7b6dd33c4, sha256 845631b5f898447de9af35bb905e5c61ec1bc20f2581f7e71ee4fc8751b129a0)
BibTeX
@misc{dcabe7b6dd33c4,
  title = {Learning In-Hand Object Reaching to General 6D Poses — arXiv:2609.13761},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13761},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_abe7b6dd33c4; sha256 845631b5f898447de9af35bb905e5c61ec1bc20f2581f7e71ee4fc8751b129a0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13761. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In-hand manipulation allows multi-fingered dexterous hands to reconfigure grasped objects without releasing and regrasping them. This improves manipulation efficiency by reducing repeated grasp acquisition and large arm motions. However, most learning-based methods focus on reorientation, continuous rotation, or translation, whereas many tasks require joint control of object position and orientation. We formulate this capability as in-hand 6D object pose reaching: starting from an existing grasp, coordinated finger motions move the object to a palm-relative target pose. We present POISE (Palm-relative Object reaching In SE(3)), a sim-to-real reinforcement learning framework for this task. POISE combines diverse stable-grasp initialization, goal- and geometry-conditioned control, an adaptive 6D goal curriculum, and a compact reward scheme for pose reaching and grasp preservation. In simulation, diverse initialization raises held-out-grasp success from 40.1% to 51.5% and post-drop recovery from 33.8% to 72.9%; the curriculum raises full-range success from 6.2% to 59.5%. On hardware, the grasp-maintenance reward improves three-target sequence success from 20% to 80%. In real-world experiments, POISE reaches successive 6D targets without manual reset across multiple object geometries and wrist orientations, and recovers from external disturbances. To support further research in dexterous manipulation, we will release our code at this https URL .

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/2609.13761 — 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 2026-10-16T01:52:32.305Z 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).