Evaluation of feature representation and machine learning methods in grasp stability learning

This paper addresses the problem of sensor-based grasping under uncertainty, specifically, the on-line estimation of grasp stability. We show that machine learning approaches can to some extent detect grasp stability from haptic pressure and finger joint information. Using data from both simulations and two real robotic hands, the paper compares different feature representations and machine learning methods to evaluate their performance in determining the grasp stability. A boosting classifier was found to perform the best of the methods tested.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[3]  Helge J. Ritter,et al.  Using entropy for dimension reduction of tactile data , 2009, 2009 International Conference on Advanced Robotics.

[4]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[5]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[7]  Heinz Wörn,et al.  Haptic object recognition using passive joints and haptic key features , 2010, 2010 IEEE International Conference on Robotics and Automation.

[8]  Danica Kragic,et al.  Learning grasp stability based on haptic data , 2010 .

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

[10]  Wolfram Burgard,et al.  Object identification with tactile sensors using bag-of-features , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[11]  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).

[12]  Javier Felip,et al.  Robust sensor-based grasp primitive for a three-finger robot hand , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[13]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[14]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[15]  Jürgen Sturm,et al.  Tactile object class and internal state recognition for mobile manipulation , 2010, 2010 IEEE International Conference on Robotics and Automation.

[16]  Masatsugu Iribe,et al.  Adaptive grasping by multi fingered hand with tactile sensor based on robust force and position control , 2008, 2008 IEEE International Conference on Robotics and Automation.

[17]  Oussama Khatib,et al.  Bayesian estimation for autonomous object manipulation based on tactile sensors , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[18]  Peter K. Allen,et al.  Examples of 3D grasp quality computations , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[19]  Dominiek Reynaerts,et al.  Featureless classification of tactile contacts in a gripper using neural networks , 1996 .