Haptic exploration for 3D shape reconstruction using five-finger hands

Abstract

In order for humanoid robots to enter human-centered environments, it is indispensable to equip them with the ability to recognize and classify objects in such an environment. A promising way to acquire the object models necessary for object manipulation appears in the supplement of the information gathered by computer vision techniques with data from haptic exploration. In this paper we present a framework for haptic exploration which is intended for use with both, a five-finger humanoid robot hand as well as with a human hand. We describe experiments and results on haptic exploration and shape estimation of 2D and 3D objects by the human hand. Volumetric shape data is acquired by a human operator hand using a data glove. The exploring human hand is located by a stereo camera system, whereas the finger configuration is calculated from the glove data.