EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 3.0

A Study on Robotic Grasping Optimization Using Point Cloud Completion Uncertainty and Grasping Stability Constraints

Young Tae Choi, Mun Haeng Lee, Sung Jae Kim, Jin-Ho Suh

Journal of the Korean Society for Precision Engineering · 01.11.2026 · 43 (11) · 1185-1193

DOI: 10.7736/jkspe.026.00066 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Korean Society for Precision Engineering, erschienen am 01.11.2026. Verfasst von Young Tae Choi und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 3.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Although 3D shape completion facilitates robotic grasping from partial observations, deterministic completion models overlook regional uncertainty, which can result in suboptimal grasp poses. To address this limitation, we propose an uncertainty-aware robotic grasping framework that couples geometric completion confidence with physical constraints. The framework integrates a Multi-rate Dropout Ensemble (MDE) into the shape completion network, enabling it to restore 3D geometry while simultaneously quantifying epistemic uncertainty. The quantified uncertainty is then mapped onto a 3D gripper bounding box to evaluate the local reliability of each grasp candidate. Evaluations in simulation on a diverse set of unseen objects demonstrate that our approach effectively suppresses unreliable grasp hypotheses. Overall, the framework significantly reduces geometric reconstruction errors and improves grasping success rates, confirming its strong generalization capability and operational safety under occlusion.

Zitieren

APA

Choi, Y. T., Lee, M. H., Kim, S. J., & Suh, J. (2026). A Study on Robotic Grasping Optimization Using Point Cloud Completion Uncertainty and Grasping Stability Constraints. Journal of the Korean Society for Precision Engineering, 43(11), 1185-1193. https://doi.org/10.7736/jkspe.026.00066

Vancouver

Choi YT, Lee MH, Kim SJ, Suh J. A Study on Robotic Grasping Optimization Using Point Cloud Completion Uncertainty and Grasping Stability Constraints. Journal of the Korean Society for Precision Engineering. 2026;43(11):1185-1193. doi:10.7736/jkspe.026.00066

BibTeX
@article{Choi2026,
  author = {Choi, Young Tae and Lee, Mun Haeng and Kim, Sung Jae and Suh, Jin-Ho},
  title = {A Study on Robotic Grasping Optimization Using Point Cloud Completion Uncertainty and Grasping Stability Constraints},
  journal = {Journal of the Korean Society for Precision Engineering},
  year = {2026},
  volume = {43},
  number = {11},
  pages = {1185--1193},
  doi = {10.7736/jkspe.026.00066}
}
Kostenlos starten →