Metraipolis › archive › dc_840c09d682cf
Iterative Grasp Pose Refinement: A Deep Reinforcement Learning Approach for 2D Vision — arXiv:2608.17628
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.17628
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T14:33:37.358Z
- SHA-256 of the archived text
04c8da9229dd7cde7fb388ddfa118fc240f3ec77e50d72d90ca4c0ef2402e7bc- Stable URL of this page
https://metraipolis.com/documento/dc_840c09d682cf
How to cite
Iterative Grasp Pose Refinement: A Deep Reinforcement Learning Approach for 2D Vision — arXiv:2608.17628 — arXiv (en), https://arxiv.org/abs/2608.17628 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_840c09d682cf (Metraipolis archive dc_840c09d682cf, sha256 04c8da9229dd7cde7fb388ddfa118fc240f3ec77e50d72d90ca4c0ef2402e7bc)
BibTeX
@misc{dc840c09d682cf,
title = {Iterative Grasp Pose Refinement: A Deep Reinforcement Learning Approach for 2D Vision — arXiv:2608.17628},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.17628},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_840c09d682cf; sha256 04c8da9229dd7cde7fb388ddfa118fc240f3ec77e50d72d90ca4c0ef2402e7bc},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.17628. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Developing robots capable of understanding and manipulating objects requires compact, interpretable, and generalizable representations. This work proposes a reinforcement learning-based framework for robotic grasp refinement, integrating keypoint-based object representations with a Deep Q-Network (DQN). Using 2D overhead images captured in a simulated environment, a geometric-based algorithm generates initial grasp candidates, which are iteratively refined by the proposed framework, transforming failed grasps into successful ones. Experiments conducted on 300 objects from the Dex-Net dataset using a UR5 manipulator demonstrate the framework's effectiveness, achieving a 100% success rate on objects previously deemed ungraspable by geometrical methods. The framework's sim-to-real transferability is further validated through physical experiments on a Delta parallel robot, where a refined grasp successfully manipulates an object that was previously ungraspable. The findings underscore the effectiveness of reinforcement learning in addressing challenges in robotic grasping, offering a scalable and adaptable solution for contact-rich manipulation tasks. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-19 in cs.AI). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Image and Video Processing (eess.IV). Pagina del lavoro: https://arxiv.org/abs/2608.17628 — 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 2026-10-18T14:27:54.445Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-19?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).