Large-scale supervised learning of the grasp robustness of surface patch pairs

The robustness of a parallel-jaw grasp can be estimated by Monte Carlo sampling of perturbations in pose and friction but this is not computationally efficient. As an alternative, we consider fast methods using large-scale supervised learning, where the input is a description of a local surface patch at each of two contact points. We train and test with disjoint subsets of a corpus of 1.66 million grasps where robustness is estimated by Monte Carlo sampling using Dex-Net 1.0. We use the BIDMach machine learning toolkit to compare the performance of two supervised learning methods: Random Forests and Deep Learning. We find that both of these methods learn to estimate grasp robustness fairly reliably in terms of Mean Absolute Error (MAE) and ROC Area Under Curve (AUC) on a held-out test set. Speedups over Monte Carlo sampling are approximately 7500x for Random Forests and 1500x for Deep Learning.

[1]  Kenneth Y. Goldberg,et al.  Bayesian grasping , 1990, Proceedings., IEEE International Conference on Robotics and Automation.

[2]  Ken Goldberg,et al.  Stochastic plans for robotic manipulation , 1991 .

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

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

[5]  Vijay Kumar,et al.  Robotic grasping and contact: a review , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[6]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[7]  Peter K. Allen,et al.  Graspit! A versatile simulator for robotic grasping , 2004, IEEE Robotics & Automation Magazine.

[8]  Paul R. Schrater,et al.  Grasping Objects with Environmentally Induced Position Uncertainty , 2009, PLoS Comput. Biol..

[9]  M. Roa,et al.  Finding locally optimum force-closure grasps , 2009 .

[10]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[11]  Ling Xu,et al.  Human-guided grasp measures improve grasp robustness on physical robot , 2010, 2010 IEEE International Conference on Robotics and Automation.

[12]  Peter K. Allen,et al.  Data-driven grasping , 2011, Auton. Robots.

[13]  Gérard Biau,et al.  Analysis of a Random Forests Model , 2010, J. Mach. Learn. Res..

[14]  Tara N. Sainath,et al.  FUNDAMENTAL TECHNOLOGIES IN MODERN SPEECH RECOGNITION Digital Object Identifier 10.1109/MSP.2012.2205597 , 2012 .

[15]  James J. Kuffner,et al.  Physically-based grasp quality evaluation under uncertainty , 2012, 2012 IEEE International Conference on Robotics and Automation.

[16]  Danica Kragic,et al.  Generalizing grasps across partly similar objects , 2012, 2012 IEEE International Conference on Robotics and Automation.

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

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

[19]  Danica Kragic,et al.  Grasp Moduli Spaces , 2013, Robotics: Science and Systems.

[20]  John Canny,et al.  BIDMach: Large-scale Learning with Zero Memory Allocation , 2013 .

[21]  Erich Elsen,et al.  Deep Speech: Scaling up end-to-end speech recognition , 2014, ArXiv.

[22]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[23]  Danica Kragic,et al.  Learning of grasp adaptation through experience and tactile sensing , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[24]  John F. Canny,et al.  Machine learning at the limit , 2015, 2015 IEEE International Conference on Big Data (Big Data).

[25]  Matei T. Ciocarlie,et al.  GP-GPIS-OPT: Grasp planning with shape uncertainty using Gaussian process implicit surfaces and Sequential Convex Programming , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

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

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

[28]  Danica Kragic,et al.  Multi-armed bandit models for 2D grasp planning with uncertainty , 2015, 2015 IEEE International Conference on Automation Science and Engineering (CASE).

[29]  Pieter Abbeel,et al.  Image Object Label 3 D CAD Model Candidate Grasps Google Object Recognition Engine Google Cloud Storage Select Feasible Grasp with Highest Success Probability Pose EstimationCamera Robots Cloud 3 D Sensor , 2014 .

[30]  Bin Fang,et al.  A comparison of 3D shape retrieval methods based on a large-scale benchmark supporting multimodal queries , 2015, Comput. Vis. Image Underst..

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

[32]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[33]  Ross A. Knepper,et al.  DeepMPC: Learning Deep Latent Features for Model Predictive Control , 2015, Robotics: Science and Systems.

[34]  Oussama Khatib,et al.  Springer Handbook of Robotics , 2007, Springer Handbooks.

[35]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[36]  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).

[37]  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).