Learning ambidextrous robot grasping policies

An ambidextrous grasping policy trained on synthetic datasets consistently clears bins of up to 25 novel objects with a physical robot. Universal picking (UP), or reliable robot grasping of a diverse range of novel objects from heaps, is a grand challenge for e-commerce order fulfillment, manufacturing, inspection, and home service robots. Optimizing the rate, reliability, and range of UP is difficult due to inherent uncertainty in sensing, control, and contact physics. This paper explores “ambidextrous” robot grasping, where two or more heterogeneous grippers are used. We present Dexterity Network (Dex-Net) 4.0, a substantial extension to previous versions of Dex-Net that learns policies for a given set of grippers by training on synthetic datasets using domain randomization with analytic models of physics and geometry. We train policies for a parallel-jaw and a vacuum-based suction cup gripper on 5 million synthetic depth images, grasps, and rewards generated from heaps of three-dimensional objects. On a physical robot with two grippers, the Dex-Net 4.0 policy consistently clears bins of up to 25 novel objects with reliability greater than 95% at a rate of more than 300 mean picks per hour.

[1]  Xinyu Liu,et al.  Dex-Net 3.0: Computing Robust Robot Suction Grasp Targets in Point Clouds using a New Analytic Model and Deep Learning , 2017, ArXiv.

[2]  Peter Corke,et al.  Closing the Loop for Robotic Grasping: A Real-time, Generative Grasp Synthesis Approach , 2018, Robotics: Science and Systems.

[3]  Máximo A. Roa,et al.  Grasp quality evaluation done right: How assumed contact force bounds affect Wrench-based quality metrics , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

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

[5]  Mark R. Cutkosky,et al.  Estimating friction using incipient slip sensing during a manipulation task , 1993, [1993] Proceedings IEEE International Conference on Robotics and Automation.

[6]  Oliver Brock,et al.  Guest Editorial Open Discussion of Robot Grasping Benchmarks, Protocols, and Metrics , 2018, IEEE Trans Autom. Sci. Eng..

[7]  Kate Saenko,et al.  Grasp Pose Detection in Point Clouds , 2017, Int. J. Robotics Res..

[8]  Kenneth Y. Goldberg,et al.  Cloud-based robot grasping with the google object recognition engine , 2013, 2013 IEEE International Conference on Robotics and Automation.

[9]  Wojciech Zaremba,et al.  Domain randomization for transferring deep neural networks from simulation to the real world , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[10]  Razvan Pascanu,et al.  Overcoming catastrophic forgetting in neural networks , 2016, Proceedings of the National Academy of Sciences.

[11]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[12]  John F. Canny,et al.  Robot Bed-Making: Deep Transfer Learning Using Depth Sensing of Deformable Fabric , 2018, ArXiv.

[13]  Stefan Leutenegger,et al.  Deep learning a grasp function for grasping under gripper pose uncertainty , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  Kate Saenko,et al.  Learning a visuomotor controller for real world robotic grasping using easily simulated depth images , 2017, ArXiv.

[15]  Kenneth Y. Goldberg,et al.  Learning Deep Policies for Robot Bin Picking by Simulating Robust Grasping Sequences , 2017, CoRL.

[16]  Kenneth Y. Goldberg,et al.  Linear Push Policies to Increase Grasp Access for Robot Bin Picking , 2018, 2018 IEEE 14th International Conference on Automation Science and Engineering (CASE).

[17]  Danica Kragic,et al.  Classical grasp quality evaluation: New algorithms and theory , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[18]  Abhinav Gupta,et al.  Robot Learning in Homes: Improving Generalization and Reducing Dataset Bias , 2018, NeurIPS.

[19]  Oliver Brock,et al.  A novel type of compliant and underactuated robotic hand for dexterous grasping , 2016, Int. J. Robotics Res..

[20]  Kenneth Y. Goldberg,et al.  Segmenting Unknown 3D Objects from Real Depth Images using Mask R-CNN Trained on Synthetic Point Clouds , 2018, ArXiv.

[21]  Gaurav S. Sukhatme,et al.  Contact localization on grasped objects using tactile sensing , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[22]  Ken Goldberg,et al.  Segmenting Unknown 3D Objects from Real Depth Images using Mask R-CNN Trained on Synthetic Data , 2018, 2019 International Conference on Robotics and Automation (ICRA).

[23]  Peter K. Allen,et al.  Pose error robust grasping from contact wrench space metrics , 2012, 2012 IEEE International Conference on Robotics and Automation.

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

[25]  Radu Bogdan Rusu,et al.  3D is here: Point Cloud Library (PCL) , 2011, 2011 IEEE International Conference on Robotics and Automation.

[26]  Robert D. Howe,et al.  Tactile sensing and control of robotic manipulation , 1993, Adv. Robotics.

[27]  Kuan-Ting Yu,et al.  A Summary of Team MIT's Approach to the Amazon Picking Challenge 2015 , 2016, ArXiv.

[28]  Maria Bauzá,et al.  Tactile Regrasp: Grasp Adjustments via Simulated Tactile Transformations , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[29]  Sergey Levine,et al.  Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[30]  John F. Canny,et al.  Part pose statistics: estimators and experiments , 1999, IEEE Trans. Robotics Autom..

[31]  Peter I. Corke,et al.  Cartman: The Low-Cost Cartesian Manipulator that Won the Amazon Robotics Challenge , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[32]  Gaurav S. Sukhatme,et al.  An autonomous manipulation system based on force control and optimization , 2014, Auton. Robots.

[33]  Andrew J. Davison,et al.  Transferring End-to-End Visuomotor Control from Simulation to Real World for a Multi-Stage Task , 2017, CoRL.

[34]  Kenneth Y. Goldberg,et al.  Design of parallel-jaw gripper tip surfaces for robust grasping , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[35]  Mathieu Aubry,et al.  Dex-Net 1.0: A cloud-based network of 3D objects for robust grasp planning using a Multi-Armed Bandit model with correlated rewards , 2016, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[36]  Ian Taylor,et al.  Robotic pick-and-place of novel objects in clutter with multi-affordance grasping and cross-domain image matching , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[37]  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).

[38]  Maciej Klimek,et al.  Improved GQ-CNN: Deep Learning Model for Planning Robust Grasps , 2018, ArXiv.

[39]  Danica Kragic,et al.  Data-Driven Grasp Synthesis—A Survey , 2013, IEEE Transactions on Robotics.

[40]  Alberto Rodriguez,et al.  Friction Variability in Planar Pushing Data: Anisotropic Friction and Data-Collection Bias , 2018, IEEE Robotics and Automation Letters.

[41]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[42]  Marcin Andrychowicz,et al.  Sim-to-Real Transfer of Robotic Control with Dynamics Randomization , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[43]  Reuven Y. Rubinstein,et al.  Fast Sequential Monte Carlo Methods for Counting and Optimization , 2013 .

[44]  Jitendra Malik,et al.  More Than a Feeling: Learning to Grasp and Regrasp Using Vision and Touch , 2018, IEEE Robotics and Automation Letters.

[45]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[46]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[47]  Sergey Levine,et al.  (CAD)$^2$RL: Real Single-Image Flight without a Single Real Image , 2016, Robotics: Science and Systems.

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

[49]  Razvan Pascanu,et al.  Sim-to-Real Robot Learning from Pixels with Progressive Nets , 2016, CoRL.

[50]  Xinyu Liu,et al.  Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics , 2017, Robotics: Science and Systems.

[51]  Ling Xu,et al.  Physical Human Interactive Guidance: Identifying Grasping Principles From Human-Planned Grasps , 2012, IEEE Transactions on Robotics.

[52]  Leslie Pack Kaelbling,et al.  Planning and Acting in Partially Observable Stochastic Domains , 1998, Artif. Intell..

[53]  Sergey Levine,et al.  Learning hand-eye coordination for robotic grasping with deep learning and large-scale data collection , 2016, Int. J. Robotics Res..

[54]  Geoffrey J. Gordon,et al.  A Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning , 2010, AISTATS.

[55]  Alberto Rodriguez,et al.  Learning Synergies Between Pushing and Grasping with Self-Supervised Deep Reinforcement Learning , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

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

[57]  Alberto Rodriguez,et al.  Reactive Planar Manipulation with Convex Hybrid MPC , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).