Grasp Pose Detection in Dense Clutter

Recently, a number of grasp detection methods have been proposed that can be used to localize robotic grasp configurations directly from sensor data without estimating object pose. The underlying idea is to treat grasp perception analogously to object detection in computer vision. These methods take as input a noisy and partially occluded RGBD image or point cloud and produce as output pose estimates of viable grasps, without assuming a known CAD model of the object. Although these methods generalize grasp knowledge to new objects well, they have not yet been demonstrated to be reliable enough to be used widely. Many grasp detection methods achieve grasp success rates (grasp successes as a fraction of the total number of grasp attempts) between 75% and 95% for novel objects presented in isolation or in light clutter. Not only are these success rates too low for practical grasping applications, but the light clutter scenarios that are evaluated often do not reflect the realities of real world grasping. This paper proposes a number of innovations that together result in a significant improvement in grasp detection performance. The specific improvement in performance due to each of our contributions is quantitatively measured either in simulation or on robotic hardware. Ultimately, we report a series of robotic experiments that average a 93% end-to-end grasp success rate for novel objects presented in dense clutter.

[1]  Ales Leonardis,et al.  One-shot learning and generation of dexterous grasps for novel objects , 2016, Int. J. Robotics Res..

[2]  Kate Saenko,et al.  High precision grasp pose detection in dense clutter , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[3]  Abhinav Gupta,et al.  Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours , 2015, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[4]  Peter K. Allen,et al.  Generating multi-fingered robotic grasps via deep learning , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[5]  Olaf Kähler,et al.  Very High Frame Rate Volumetric Integration of Depth Images on Mobile Devices , 2015, IEEE Transactions on Visualization and Computer Graphics.

[6]  Jeannette Bohg,et al.  Leveraging big data for grasp planning , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[7]  Subhransu Maji,et al.  Multi-view Convolutional Neural Networks for 3D Shape Recognition , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[8]  Joseph Redmon,et al.  Real-time grasp detection using convolutional neural networks , 2014, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[9]  Honglak Lee,et al.  Deep learning for detecting robotic grasps , 2013, Int. J. Robotics Res..

[10]  Robert Platt,et al.  Using Geometry to Detect Grasp Poses in 3D Point Clouds , 2015, ISRR.

[11]  Florian Schmidt,et al.  Learning dexterous grasps that generalise to novel objects by combining hand and contact models , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[12]  Pieter Abbeel,et al.  BigBIRD: A large-scale 3D database of object instances , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[13]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[14]  Pieter Abbeel,et al.  Finding Locally Optimal, Collision-Free Trajectories with Sequential Convex Optimization , 2013, Robotics: Science and Systems.

[15]  Markus Vincze,et al.  Learning grasps for unknown objects in cluttered scenes , 2013, 2013 IEEE International Conference on Robotics and Automation.

[16]  Danica Kragic,et al.  Learning a dictionary of prototypical grasp-predicting parts from grasping experience , 2013, 2013 IEEE International Conference on Robotics and Automation.

[17]  Jared Glover,et al.  Bingham procrustean alignment for object detection in clutter , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[18]  Markus Vincze,et al.  Empty the basket - a shape based learning approach for grasping piles of unknown objects , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[19]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[20]  Alexander Herzog,et al.  Template-based learning of grasp selection , 2012, 2012 IEEE International Conference on Robotics and Automation.

[21]  Oliver Kroemer,et al.  A kernel-based approach to direct action perception , 2012, 2012 IEEE International Conference on Robotics and Automation.

[22]  Markus Vincze,et al.  3DNet: Large-scale object class recognition from CAD models , 2012, 2012 IEEE International Conference on Robotics and Automation.

[23]  S. Chitta,et al.  Perception , Planning , and Execution for Mobile Manipulation in Unstructured Environments , 2012 .

[24]  Ashutosh Saxena,et al.  Efficient grasping from RGBD images: Learning using a new rectangle representation , 2011, 2011 IEEE International Conference on Robotics and Automation.

[25]  Takeo Kanade,et al.  Automated Construction of Robotic Manipulation Programs , 2010 .

[26]  Ashutosh Saxena,et al.  Robotic Grasping of Novel Objects using Vision , 2008, Int. J. Robotics Res..

[27]  P. Allen,et al.  Dexterous Grasping via Eigengrasps : A Low-dimensional Approach to a High-complexity Problem , 2007 .

[28]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[29]  Richard M. Murray,et al.  A Mathematical Introduction to Robotic Manipulation , 1994 .

[30]  John F. Canny,et al.  Planning optimal grasps , 1992, Proceedings 1992 IEEE International Conference on Robotics and Automation.