Abstract
Grasping objects poses a significant challenge for autonomous robotic manipulation in unstructured and cluttered environments. Despite recent advancements in 6-DoF (degree of freedom) grasp pose estimation, the majority of existing methods fail to differentiate between points on adjacent objects, particularly in scenarios where objects are positioned next to each other. This imprecise orientation often results in collisions or unsuccessful grasps. To address these challenges, this paper proposes a semantic instance reconstruction grasp network (SIRGN) that efficiently generates accurate grasping configurations. Firstly, the foreground objects are reconstructed using the implicit semantic instance branch. Through voting, we predict the corresponding instance for each foreground point and thereby differentiate adjacent objects. Secondly, to enhance the accuracy of grasping orientation, we decompose the 3D rotation matrix into two orthogonal unit vectors. Furthermore, the network is trained using VGN simulation grasping datasets. The results of declutter experiments demonstrate that the grasp success rate of SIRGN in packed and pile scenes is 89.5% and 78.1%, respectively. Experiments conducted in both simulated and real environments have fully demonstrated the effectiveness of the proposed methodology.
Similar content being viewed by others
References
Li G, Liang X, Gao Y, Su T, Liu Z, Hou ZG (2023) A linkage-driven underactuated robotic hand for adaptive grasping and in-hand manipulation. IEEE Trans Autom Sci Eng. https://doi.org/10.1109/TASE.2023.3273721
Zhu H, Li X, Chen W, Li X, Ma J, Teo CS, Lin W (2022) Weight imprinting classification-based force grasping with a variable-stiffness robotic gripper. IEEE Trans Autom Sci Eng 19(2):961–981. https://doi.org/10.1109/TASE.2021.3054655
Newbury R, Gu M, Chumbley L, Mousavian A, Eppner C, Leitner J, Bohg J, Morales A, Asfour T, Kragic D, Fox D, Cosgun A (2022) A deep learning approaches to grasp synthesis: a review. arXiv preprint arXiv:2207.02556. https://doi.org/10.48550/arXiv.2207.02556
Zhang Y, Yang Q (2022) A survey on multi-task learning. IEEE Trans Knowl Data Eng 34(12):5586–5609. https://doi.org/10.1109/TKDE.2021.3070203
Du G, Wang K, Lian S, Zhao K (2021) Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review. Artif Intell Rev 54(3):1677–1734. https://doi.org/10.1007/s10462-020-09888-5
Chiu MC, Tsai HY, Chiu JE (2022) A novel directional object detection method for piled objects using a hybrid region-based convolutional neural network. Adv Eng Inform 51:101448. https://doi.org/10.1016/j.aei.2021.101448
Zhu X, Wang D, Biza O, Su G, Walters R, Platt R (2022) Sample efficient grasp learning using equivariant models. arXiv preprint arXiv:2202.09468. https://doi.org/10.48550/arXiv.2202.09468
Yu X, Zhuang Z, Koniusz P, Li H (2020) 6DoF object pose estimation via differentiable proxy voting loss. arXiv preprint arXiv:2002.03923. https://doi.org/10.48550/arXiv.2002.03923
Tian H, Song K, Li S, Ma S, Xu J, Yan Y (2023) Data-driven robotic visual grasping detection for unknown objects: a problem-oriented review. Expert Syst Appl 211:118624. https://doi.org/10.1016/j.eswa.2022.118624
Zhao S, Li B, Xu P, Keutzer K (2020) Multi-source domain adaptation in the deep learning era: a systematic survey. arXiv preprint arXiv:2002.12169. https://doi.org/10.48550/arXiv.2002.12169
Hong QQ, Yang L, Zeng B (2022) RANET: a grasp generative residual attention network for robotic grasping detection. Int J Control Autom Syst 20:3996–4004. https://doi.org/10.1007/s12555-021-0929-8
Wu C, Chen J, Cao Q, Zhang J, Tai Y, Sun L, Jia K (2020) Grasp proposal networks: an end-to-end solution for visual learning of robotic grasps. Adv Neural Inf Process Syst 33:13174–13184
Suzuki Y, Yoshida R, Tsuji T, Nishimura T, Watanabe T (2022) Grasping strategy for unknown objects based on real-time grasp-stability evaluation using proximity sensing. IEEE Robot Autom Lett 7(4):8643–8650. https://doi.org/10.1109/LRA.2022.3188885
Wang JW, Li CL, Chen JL, Lee JJ (2022) Robot grasping in dense clutter via view-based experience transfer. Int J Intell Robot Appl 6:23–37. https://doi.org/10.1007/s41315-021-00179-y
Lilge S, Barfoot TD, Burgner-Kahrs J (2022) Continuum robot state estimation using Gaussian process regression on SE (3). Int J Robot Res 41(13–14):1099–1120. https://doi.org/10.1177/02783649221128843
Bohg J, Morales A, Asfour T, Kragic D (2013) Data-driven grasp synthesis—a survey. IEEE Trans Robot 30(2):289–309. https://doi.org/10.1109/TRO.2013.2289018
Tian H, Song K, Li S, Ma S, Yan Y (2023) Rotation adaptive grasping estimation network oriented to unknown objects based on novel RGB-D fusion strategy. Eng Appl Artif Intell 120:105842. https://doi.org/10.1016/j.engappai.2023.105842
Kim S, Ahn T, Lee Y, Kim J, Wang MY, Park FC (2022) DSQNet: a deformable model-based supervised learning algorithm for grasping unknown occluded objects. IEEE Trans Autom Sci Eng 20(3):1721–1734. https://doi.org/10.1109/TASE.2022.3184873
Morrison D, Corke P, Leitner J (2018) Closing the loop for robotic grasping: a real-time, generative grasp synthesis approach. arXiv preprint arXiv:1804.05172. https://doi.org/10.48550/arXiv.1804.05172
Sayour MH, Kozhaya SE, Saab SS (2022) Autonomous robotic manipulation: real-time, deep-learning approach for grasping of unknown objects. J Robot 2022:2585656. https://doi.org/10.1155/2022/2585656
Fang K, Zhu Y, Garg A, Kurenkov A, Mehta V, Fei-Fei L, Savarese S (2020) Learning task-oriented grasping for tool manipulation from simulated self-supervision. Int J Robot Res 39(2–3):202–216. https://doi.org/10.1177/0278364919872545
Mahler J, Liang J, Niyaz S, Laskey M, Doan R, Liu X, Goldberg K (2017) Dex-net 2.0: deep learning to plan robust grasps with synthetic point clouds and analytic grasp metrics. arXiv preprint arXiv:1703.09312. https://doi.org/10.48550/arXiv.1703.09312
Zeng A, Song S, Yu KT, Donlon E, Hogan FR, Bauza M, Rodriguez A (2022) Robotic pick-and-place of novel objects in clutter with multi-affordance grasping and cross-domain image matching. Int J Robot Res 41(7):690–705. https://doi.org/10.1177/0278364919868017
Mousavian A, Eppner C, Fox D (2019) 6-DoF graspnet: variational grasp generation for object manipulation. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 2901–2910
Ten Pas A, Gualtieri M, Saenko K, Platt R (2017) Grasp pose detection in point clouds. Int J Robot Res 36(13–14):1455–1473. https://doi.org/10.1177/0278364917735594
Liang H, Ma X, Li S, Görner M, Tang S, Fang B, Zhang J (2019) Pointnetgpd: detecting grasp configurations from point sets. In: 2019 International conference on robotics and automation (ICRA), Montreal, IEEE, pp 3629–3635. https://doi.org/10.1109/ICRA.2019.8794435
Breyer M, Chung JJ, Ott L, Siegwart R, Nieto J (2021) Volumetric grasping network: real-time 6 DoF grasp detection in clutter. In: Proceedings of the 2020 conference on robot learning, PMLR, vol 155, pp 1602–1611.
Li G, Xu P, Qiao S, Li B (2021) Stability analysis and optimal enveloping grasp planning of a deployable robotic hand. Mech Mach Theory 158:104241. https://doi.org/10.1016/j.mechmachtheory.2020.104241
Chen S, Tang W, Xie P, Yang W, Wang G (2023) Efficient heatmap-guided 6-DoF grasp detection in cluttered scenes. IEEE Robot Autom Lett 8(8):4895–4902. https://doi.org/10.1109/LRA.2023.3290513
Yan X, Hsu J, Khansari M, Bai Y, Pathak A, Gupta A, Lee H (2018) Learning 6-DoF grasping interaction via deep geometry-aware 3d representations. In: 2018 IEEE international conference on robotics and automation (ICRA), Brisbane, QLD, Australia, IEEE, pp 3766–3773. https://doi.org/10.1109/ICRA.2018.8460609
Li Y, Kong T, Chu R, Li Y, Wang P, Li L (2021) Simultaneous semantic and collision learning for 6-DoF grasp pose estimation. arXiv preprint arXiv:2108.02425. https://doi.org/10.48550/arXiv.2108.02425
Van der Merwe M, Lu Q, Sundaralingam B, Matak M, Hermans T (2020) Learning continuous 3d reconstructions for geometrically aware grasping. In: 2020 IEEE International conference on robotics and automation (ICRA), Paris, France, IEEE, pp 11516–11522. https://doi.org/10.1109/ICRA40945.2020.9196981
Jiang Z, Zhu Y, Svetlik M, Fang K, Zhu Y (2021) Synergies between affordance and geometry: 6-DoF grasp detection via implicit representations. arXiv preprint arXiv:2104.01542. https://doi.org/10.48550/arXiv.2104.01542
Chen B, Zhang T, Cong L, Ma J, Hu W (2022) Forward kinematics of body posture perception using an improved BP neural network based on a quantum genetic algorithm. Laser Phys Lett 19(9):095201. https://doi.org/10.1088/1612-202X/ac7f37
Fang HS, Wang C, Gou M, Lu C (2020) Graspnet-1billion: a large-scale benchmark for general object grasping. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 11444–11453
Peng S, Niemeyer M, Mescheder L, Pollefeys M, Geiger A (2020) Convolutional occupancy networks. arXiv preprint arXiv:2003.04618. https://doi.org/10.48550/arXiv.2003.04618
Grassmann R, Burgner-Kahrs J (2019) On the merits of joint space and orientation representations in learning the forward kinematics in SE (3). In: Robotics: Science and Systems. Freiburg in Breisgau, June 22–26
Danielczuk M, Matl M, Gupta S, Li A, Lee A, Mahler J, Goldberg K (2019) Segmenting unknown 3d objects from real depth images using mask R-CNN trained on synthetic data. In: 2019 International conference on robotics and automation (ICRA). Montreal, QC, Canada, IEEE, pp 7283–7290. https://doi.org/10.1109/ICRA.2019.8793744
Han L, Zheng T, Xu L, Fang L (2020) Occuseg: occupancy-aware 3d instance segmentation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 2940–2949
Xie C, Xiang Y, Mousavian A, Fox D (2021) Unseen object instance segmentation for robotic environments. IEEE Trans Robot 37(5):1343–1359. https://doi.org/10.1109/TRO.2021.3060341
Acknowledgements
The work was supported by the National Nature Science Foundation of China (62272426, 62106238) and Shanxi Province Nature Science Foundation of China (202303021211153, 202203021212138).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflicts of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Han, H., Wang, W., Han, X. et al. 6-DoF grasp pose estimation based on instance reconstruction. Intel Serv Robotics 17, 251–264 (2024). https://doi.org/10.1007/s11370-023-00489-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11370-023-00489-z