Abstract
Grasping is the first step in most robotic manipulation tasks, and it is essential for applications of robots in real-life scenarios. For humans, grasping novel objects is a naturally gained ability, however, for robots, it is a challenging task due to complex object shapes and incomplete visual information. Many current grasp pose estimation methods need to first construct 3D models of the scene and generates a large pool of grasp candidates, and then perform a search for the best grasp. These methods rely on high quality 3D models, and their long pipeline makes them unfeasible for real-time processing. End-to-end grasp pose estimation methods mitigate these issues, but they can only deals with few DoF planar grasps that fail to cover many successful grasps. In this paper, we propose a viewing angle generative network (VAGN), an approach that bridges the aforementioned two main classes of methods. VAGN decouples 7-DoF grasp detection into two stages. In the first stage, it predicts the camera viewing angle, which is also the orientation of the gripper around the object from an RGBD frame. In the second stage, it generates a planar grasp pose by taking another RGBD image at the predicted viewing angle in stage 1. We trained VAGN on the Cornell dataset. Real robot experiments on a UR-10e robot with camera-in-hand show real-time processing speed and higher success rates compared to the state-of-the-art GR-ConvNet, in both single object scenes and cluttered scenes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bicchi, A., Kumar, V.: Robotic grasping and contact: a review. In: Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation, vol. 1, pp. 348–353 (2000)
Chu, F., Xu, R., Vela, P.A.: Real-world multiobject, multigrasp detection. IEEE Robot. Autom. Lett. 3(4), 3355–3362 (2018)
Depierre, A., Dellandrea, E., Chen, L.: Jacquard: a large scale dataset for robotic grasp detection. In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3511–3516 (2018)
Detry, R., et al.: Learning object-specific grasp affordance densities. In: 2009 IEEE 8th International Conference on Development and Learning, pp. 1–7 (2009)
Fang, H.S., Wang, C., Gou, M., Lu, C.: Graspnet-1billion: A large-scale benchmark for general object grasping. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 11444–11453 (2020)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)
Herzog, A., Pastor, P., Kalakrishnan, M., Righetti, L., Bohg, J., Asfour, T., Schaal, S.: Learning of grasp selection based on shape-templates. Autonomous Robots 36, January 2014
Kingma, D., Ba, J.: Adam: A method for stochastic optimization. In: International Conference on Learning Representations (2015)
Kumra, S., Joshi, S., Sahin, F.: Antipodal robotic grasping using generative residual convolutional neural network. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), September 2020
Morrison, D., Corke, P., Leitner, J.: Learning robust, real-time, reactive robotic grasping. Int. J. Robot. Res. 39(2–3), 183–201 (2020)
Mousavian, A., Eppner, C., Fox, D.: 6-dof graspnet: variational grasp generation for object manipulation. In: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 2901–2910, May 2019
Ni, P., Zhang, W., Zhu, X., Cao, Q.: Pointnet++ grasping: learning an end-to-end spatial grasp generation algorithm from sparse point clouds. In: 2020 IEEE International Conference on Robotics and Automation (ICRA), pp. 3619–3625, March 2020
ten Pas, A., Gualtieri, M., Saenko, K., Platt, R.: Grasp pose detection in point clouds. Int. J. Robot. Res. 36(13–14), 1455–1473 (2017)
Peng, S., Zhou, X., Liu, Y., Lin, H., Huang, Q., Bao, H.: Pvnet: pixel-wise voting network for 6dof object pose estimation. IEEE Trans. Pattern Anal. Mach. Intell., 1 (2020)
Yun Jiang, Moseson, S., Saxena, A.: Efficient grasping from rgbd images: Learning using a new rectangle representation. In: 2011 IEEE International Conference on Robotics and Automation (ICRA), pp. 3304–3311 (2011)
Zeng, A., et al.: Multi-view self-supervised deep learning for 6d pose estimation in the amazon picking challenge. In: 2017 IEEE International Conference on Robotics and Automation (ICRA), pp. 1386–1383 (2017)
Zhou, X., Lan, X., Zhang, H., Tian, Z., Zhang, Y., Zheng, N.: Fully convolutional grasp detection network with oriented anchor box. In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 7223–7230 (2018)
Acknowledgement
This study was supported by Jihua Laboratory through the Self-Programming Intelligent Robot Project (No. X190101TB190) and Funds for Young Scholar (No. X201181XB200), also by Guangdong Basic and Applied Basic Research Foundation (No. 2020A1515110267).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Gao, X., Li, W., Wen, Z. (2021). Viewing Angle Generative Model for 7-DoF Robotic Grasping. In: Fang, L., Chen, Y., Zhai, G., Wang, J., Wang, R., Dong, W. (eds) Artificial Intelligence. CICAI 2021. Lecture Notes in Computer Science(), vol 13070. Springer, Cham. https://doi.org/10.1007/978-3-030-93049-3_27
Download citation
DOI: https://doi.org/10.1007/978-3-030-93049-3_27
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93048-6
Online ISBN: 978-3-030-93049-3
eBook Packages: Computer ScienceComputer Science (R0)