Self-supervised regrasping using spatio-temporal tactile features and reinforcement learning

We introduce a framework for learning regrasping behaviors based on tactile data. First, we present a grasp stability predictor that uses spatio-temporal tactile features collected from the early-object-lifting phase to predict the grasp outcome with a high accuracy. Next, the trained predictor is used to supervise and provide feedback to a reinforcement learning algorithm that learns the required grasp adjustments based on tactile feedback. Our results gathered over more than 50 hours of real robot experiments indicate that the robot is able to predict the grasp outcome with 93% accuracy. In addition, the robot is able to improve the grasp success rate from 42% when randomly grasping an object to up to 97% when allowed to regrasp the object in case of a predicted failure.

[1]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[2]  Manuel Lopes,et al.  Active learning of visual descriptors for grasping using non-parametric smoothed beta distributions , 2012, Robotics Auton. Syst..

[3]  Danica Kragic,et al.  Learning grasp stability based on tactile data and HMMs , 2010, 19th International Symposium in Robot and Human Interactive Communication.

[4]  Ian T. Jolliffe,et al.  Principal Component Analysis , 2002, International Encyclopedia of Statistical Science.

[5]  Jan Peters,et al.  Stabilizing novel objects by learning to predict tactile slip , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[6]  Peter K. Allen,et al.  Stable grasping under pose uncertainty using tactile feedback , 2014, Auton. Robots.

[7]  Jan Peters,et al.  Reinforcement Learning to Adjust Robot Movements to New Situations , 2010, IJCAI.

[8]  Oliver Kroemer,et al.  Combining active learning and reactive control for robot grasping , 2010, Robotics Auton. Syst..

[9]  Yasemin Altun,et al.  Relative Entropy Policy Search , 2010 .

[10]  Jimmy A. Jørgensen,et al.  Assessing Grasp Stability Based on Learning and Haptic Data , 2011, IEEE Transactions on Robotics.

[11]  Danica Kragic,et al.  ST-HMP: Unsupervised Spatio-Temporal feature learning for tactile data , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[12]  Gaurav S. Sukhatme,et al.  Force estimation and slip detection/classification for grip control using a biomimetic tactile sensor , 2015, 2015 IEEE-RAS 15th International Conference on Humanoid Robots (Humanoids).

[13]  Jan Peters,et al.  A Survey on Policy Search for Robotics , 2013, Found. Trends Robotics.

[14]  Veronica J. Santos,et al.  Biomimetic Tactile Sensor Array , 2008, Adv. Robotics.

[15]  Gaurav S. Sukhatme,et al.  BiGS: BioTac Grasp Stability Dataset , 2016 .

[16]  Dieter Fox,et al.  Hierarchical Matching Pursuit for Image Classification: Architecture and Fast Algorithms , 2011, NIPS.

[17]  G.E. Loeb,et al.  Grip Control Using Biomimetic Tactile Sensing Systems , 2009, IEEE/ASME Transactions on Mechatronics.

[18]  Dieter Fox,et al.  Multipath Sparse Coding Using Hierarchical Matching Pursuit , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

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

[20]  A. Bruckstein,et al.  K-SVD : An Algorithm for Designing of Overcomplete Dictionaries for Sparse Representation , 2005 .

[21]  Stefan Schaal,et al.  Skill learning and task outcome prediction for manipulation , 2011, 2011 IEEE International Conference on Robotics and Automation.

[22]  Oliver Kroemer,et al.  Learning robot tactile sensing for object manipulation , 2014, 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems.