A study on tactile texture recognition using magnetic type tactile sensor

For the purpose of developing a human-friendly robot, a texture recognition method using tactile sensing in which a magnetic type tactile sensor is used is proposed. The magnetic sensor has soft and elastic structure part which interact with a object and two sensing elements. One is GMR (Giant MagnetoResistance) which can sense displacement of the elastic part. The other is IND (INDuctor) which can sense motion and vibration of elastic part. Support Vector Machine is used to learn texture information. Through some experiments, it is find that each of two sensing elements have strong points in different frequency ranges. Then, it is showed that leaning setting in which adequate frequencies are used in GMR and IND brings high recognition rate.

[1]  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.

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

[3]  C. Spence,et al.  Skin and Touch , 2014 .

[4]  Satoru Takenawa,et al.  A magnetic type tactile sensor using a two-dimensional array of inductors , 2009, 2009 IEEE International Conference on Robotics and Automation.

[5]  Hiroyuki Nakamoto,et al.  Structure and fundamental evaluation of magnetic type tactile sensor , 2012 .