Detection and Identification of Object Based on a Magnetostrictive Tactile Sensing System

Abstract

Tactile sensing is important for the exploration and manipulation of an object. In this paper, a magnetostrictive tactile sensing system has been found. A robotic gripper with one sensor mounted on its fingers performs a palpation procedure on a set of objects. By gripping an object, the robot actively explores the material properties, and the system acquires tactile information corresponding to the pressure and stiffness. Based on the extreme learning machine classifier algorithm, taking the output voltage rising gradient and steady-state voltage as the eigenvalues, an object is detected and identified. The complex matrixes at the best classification accuracy are plotted to show the identification rate of each object. This magnetostrictive tactile sensing system implements real-time operation and the object can be identified during the grasping process. It means that the system is able to identify object successfully.