EUVIMEDEUVIMED entdecken
Crossref

Learning Part-Aware Functional Handovers for Human-to-Robot Collaboration

Binghui Zuo, Zhuolin Deng, Zhipeng Yu, Yuzhou Zheng, Wenqian Sun, Yangang Wang

IEEE Transactions on Industrial Electronics · 2026-08 · 73 (8) · 11835-11847

DOI: 10.1109/tie.2026.3670245 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Industrial Electronics, 2026-08. Verfasst von Binghui Zuo und 5 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province, Postgraduate Research & Practice Innovation Program of Jiangsu Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Human-to-robot collaboration plays a pivotal role in industrial manufacturing, where functional awareness and safety are essential. Although recent works advance generalizable solutions, they often neglect the functional characteristics of objects. In this article, we propose a part-aware handover framework, enabling the robot to recognize the intrinsic properties of objects and perform functionally aligned receiving behaviors based on human delivery actions. To support this insight, we construct a large-scale human demonstration dataset, containing over 100 times more demonstrations than mainstream benchmarks. Building upon it, we further propose a generative model to synthesize robot grasps, which is combined with two filtering criteria. A part-aware criterion enforces functional alignment between human delivering and robot receiving regions, and a safety-based criterion excludes unsafe grasps. These two criteria jointly ensure that the synthesized robot poses are both functionally appropriate and physically executable during close human-to-robot collaborations. We also train a handover policy to evaluate the effectiveness of this framework. Comprehensive experiments confirm that our method achieves superior performance in both simulated and real-world environments.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Jiangsu Province
  • Postgraduate Research & Practice Innovation Program of Jiangsu Province

Zitieren

APA

Zuo, B., Deng, Z., Yu, Z., Zheng, Y., Sun, W., & Wang, Y. (2026). Learning Part-Aware Functional Handovers for Human-to-Robot Collaboration. IEEE Transactions on Industrial Electronics, 73(8), 11835-11847. https://doi.org/10.1109/tie.2026.3670245

Vancouver

Zuo B, Deng Z, Yu Z, Zheng Y, Sun W, Wang Y. Learning Part-Aware Functional Handovers for Human-to-Robot Collaboration. IEEE Transactions on Industrial Electronics. 2026;73(8):11835-11847. doi:10.1109/tie.2026.3670245

BibTeX
@article{Zuo2026,
  author = {Zuo, Binghui and Deng, Zhuolin and Yu, Zhipeng and Zheng, Yuzhou and Sun, Wenqian and Wang, Yangang},
  title = {Learning Part-Aware Functional Handovers for Human-to-Robot Collaboration},
  journal = {IEEE Transactions on Industrial Electronics},
  year = {2026},
  volume = {73},
  number = {8},
  pages = {11835--11847},
  doi = {10.1109/tie.2026.3670245}
}
Kostenlos starten →