Volumetric-based Contact Point Detection for 7-DoF Grasping

Junhao Cai, Jingcheng Su, Zida Zhou, Hui Cheng, Qifeng Chen, Michael Y Wang
Proceedings of The 6th Conference on Robot Learning, PMLR 205:824-834, 2023.

Abstract

In this paper, we propose a novel grasp pipeline based on contact point detection on the truncated signed distance function (TSDF) volume to achieve closed-loop 7-degree-of-freedom (7-DoF) grasping on cluttered environments. The key aspects of our method are that 1) the proposed pipeline exploits the TSDF volume in terms of multi-view fusion, contact-point sampling and evaluation, and collision checking, which provides reliable and collision-free 7-DoF gripper poses with real-time performance; 2) the contact-based pose representation effectively eliminates the ambiguity introduced by the normal-based methods, which provides a more precise and flexible solution. Extensive simulated and real-robot experiments demonstrate that the proposed pipeline can select more antipodal and stable grasp poses and outperforms normal-based baselines in terms of the grasp success rate in both simulated and physical scenarios. Code and data are available at https://github.com/caijunhao/vcpd

Cite this Paper

BibTeX
@InProceedings{pmlr-v205-cai23a, title = {Volumetric-based Contact Point Detection for 7-DoF Grasping}, author = {Cai, Junhao and Su, Jingcheng and Zhou, Zida and Cheng, Hui and Chen, Qifeng and Wang, Michael Y}, booktitle = {Proceedings of The 6th Conference on Robot Learning}, pages = {824--834}, year = {2023}, editor = {Liu, Karen and Kulic, Dana and Ichnowski, Jeff}, volume = {205}, series = {Proceedings of Machine Learning Research}, month = {14--18 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v205/cai23a/cai23a.pdf}, url = {https://proceedings.mlr.press/v205/cai23a.html}, abstract = {In this paper, we propose a novel grasp pipeline based on contact point detection on the truncated signed distance function (TSDF) volume to achieve closed-loop 7-degree-of-freedom (7-DoF) grasping on cluttered environments. The key aspects of our method are that 1) the proposed pipeline exploits the TSDF volume in terms of multi-view fusion, contact-point sampling and evaluation, and collision checking, which provides reliable and collision-free 7-DoF gripper poses with real-time performance; 2) the contact-based pose representation effectively eliminates the ambiguity introduced by the normal-based methods, which provides a more precise and flexible solution. Extensive simulated and real-robot experiments demonstrate that the proposed pipeline can select more antipodal and stable grasp poses and outperforms normal-based baselines in terms of the grasp success rate in both simulated and physical scenarios. Code and data are available at https://github.com/caijunhao/vcpd} }
Endnote
%0 Conference Paper %T Volumetric-based Contact Point Detection for 7-DoF Grasping %A Junhao Cai %A Jingcheng Su %A Zida Zhou %A Hui Cheng %A Qifeng Chen %A Michael Y Wang %B Proceedings of The 6th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Karen Liu %E Dana Kulic %E Jeff Ichnowski %F pmlr-v205-cai23a %I PMLR %P 824--834 %U https://proceedings.mlr.press/v205/cai23a.html %V 205 %X In this paper, we propose a novel grasp pipeline based on contact point detection on the truncated signed distance function (TSDF) volume to achieve closed-loop 7-degree-of-freedom (7-DoF) grasping on cluttered environments. The key aspects of our method are that 1) the proposed pipeline exploits the TSDF volume in terms of multi-view fusion, contact-point sampling and evaluation, and collision checking, which provides reliable and collision-free 7-DoF gripper poses with real-time performance; 2) the contact-based pose representation effectively eliminates the ambiguity introduced by the normal-based methods, which provides a more precise and flexible solution. Extensive simulated and real-robot experiments demonstrate that the proposed pipeline can select more antipodal and stable grasp poses and outperforms normal-based baselines in terms of the grasp success rate in both simulated and physical scenarios. Code and data are available at https://github.com/caijunhao/vcpd
APA
Cai, J., Su, J., Zhou, Z., Cheng, H., Chen, Q. & Wang, M.Y.. (2023). Volumetric-based Contact Point Detection for 7-DoF Grasping. Proceedings of The 6th Conference on Robot Learning, in Proceedings of Machine Learning Research 205:824-834 Available from https://proceedings.mlr.press/v205/cai23a.html.

Related Material