Metraipolis › archive › dc_aabe628e0031
Hybrid Residual Reinforcement Learning for Contact-Rich Robotic Book Insertion — arXiv:2609.19962
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.19962
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T05:25:37.175Z
- SHA-256 of the archived text
ad6c466c09f0b1560540294e7668cedbd93c5db50bbf0705026970931fe4b4db- Stable URL of this page
https://metraipolis.com/documento/dc_aabe628e0031
How to cite
Hybrid Residual Reinforcement Learning for Contact-Rich Robotic Book Insertion — arXiv:2609.19962 — arXiv (en), https://arxiv.org/abs/2609.19962 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aabe628e0031 (Metraipolis archive dc_aabe628e0031, sha256 ad6c466c09f0b1560540294e7668cedbd93c5db50bbf0705026970931fe4b4db)
BibTeX
@misc{dcaabe628e0031,
title = {Hybrid Residual Reinforcement Learning for Contact-Rich Robotic Book Insertion — arXiv:2609.19962},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.19962},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aabe628e0031; sha256 ad6c466c09f0b1560540294e7668cedbd93c5db50bbf0705026970931fe4b4db},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19962. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Placing a grasped book into a tight shelf is a compact but difficult contact-rich control problem: millimetre-scale pose error can turn a geometrically valid approach into jamming, failed release, or incomplete seating. We study this final phase after grasp acquisition and global approach, and ask how control authority should be divided between known geometry and learned behaviour. Our method retains a nominal task-space controller for structured insertion and seating, while residual PPO supplies bounded local corrections and decides when to release. Only the brief open-retreat-reclose transition is scripted. For the final policy used on hardware, a deployment-matched simulation evaluation over 512 fixed conditions yields 98.50 percent mean success (0.23 percentage-point sample SD) across three independent training runs, compared with 37.89 percent for nominal control. On the physical xArm7, 60 trials over 30 matched conditions show the same qualitative advantage: residual control raises success from 26.7 percent to 63.3 percent, reduces failures from 22 to 11, and wins 13 of the 15 matched conditions in which the two controllers differ. Robustness tests show that performance remains above 87 percent under initialization perturbations up to 1.5x, while very tight clearances expose the geometric limit of local correction. These results support a hybrid design in which geometry preserves reliable task structure and learning is concentrated on the contact-sensitive behaviour that fixed rules handle poorly. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-18 in cs.RO). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.19962 — 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 2026-10-10T04:47:38.868Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).