科学技术与应用

  • Home
  • About
    • About the Journal
    • Contact
  • Article
    • Current
    • Archives
  • Submissions
  • Editorial Team
  • Announcements
Register Login

ISSN

3060-9453(Oline)
3060-9461(Print)

Article Processing Charges (APCs)

SGD$600

Publication Frequency

Bi-Monthly

PDF全文下载

Published

2026-08-04

Issue

Vol 3 No 3 (2026): Published

Section

Articles

How to Cite

  • ACM
  • ACS
  • APA
  • ABNT
  • Chicago
  • Harvard
  • IEEE
  • MLA
  • Turabian
  • Vancouver

  • Download Citation
  • Endnote/Zotero/Mendeley (RIS)
  • BibTeX

基于 Point Transformer 的条件扩散六自由度位姿生成

雷 啸天

北方工业大学人工智能与计算机学院/大规模流数据集成与分析技术北京市重点实验室

王 菁

北方工业大学人工智能与计算机学院/大规模流数据集成与分析技术北京市重点实验室


DOI: https://doi.org/10.59429/kxjsyy.v3i3.14797


Keywords: 六自由度抓取;多模态融合;交叉注意力


Abstract

随着深度学习的发展,基于点云的六自由度抓取感知与生成方法取得了显著进展。传统主要侧重于几何与物 理稳定性评价,在面向自然语言指令驱动的抓取任务时,难以在候选生成阶段有效融入语义约束。为此,本文提出 一种基于 Point Transformer 条件扩散六自由度抓取生成方法,利用 Point Transformer 对场景点云进行特征提取,在条 件扩散去噪网络中引入交叉注意力机制,实现语言指令与点云特征的深层对齐与融合,进而在连续抓取空间中生成 任务相关的抓取候选,设计联合一致性评分对候选抓取进行评估与筛选,在保证物理可执行性的同时提升语义匹配 性与抓取完成质量。实验结果表明,该方法在语义一致性与抓取成功率等指标上优于对比方法,具备良好的泛化能 力与应用潜力。


References

[1] Suárez-Ruiz F, Zhou X, Pham Q C. Can robots assemble an IKEA chair?[J]. Science Robotics, 2018, 3(17):eaat6385.

[2] Dantam N T, Kingston Z K, Chaudhuri S, et al.Incremental task and motion planning: A constraint-based approach[C]//Robotics: Science and systems. 2016, 12: 00052.

[3] Ten Pas A, Gualtieri M, Saenko K, et al. Grasp pose detection in point clouds[J]. The International Journal of Robotics Research, 2017, 36(13-14): 1455-1473.

[4] Liang H, Ma X, Li S, et al. Pointnetgpd: Detecting grasp configurations from point sets[C]//2019 international conference on robotics and automation (ICRA). IEEE, 2019:3629-3635.

[5] Fang H S, Wang C, Gou M, et al. Graspnet-1billion: A large-scale benchmark for general object grasping[C]// Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 11444-11453.

[6] Li M, Zhao Q, Lyu S, et al. Ovgnet: A unified visual-linguistic framework for open-vocabulary robotic grasping[C]//2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2024: 7507-7513.

[7] Achiam J, Adler S, Agarwal S, et al. Gpt-4 technical report[J]. arXiv preprint arXiv:2303.08774, 2023.



ISSN: 3060-9453
21 Woodlands Close #02-10 Primz Bizhub Singapore 737854

Email:editorial_office@as-pub.com