Skip to Main Content
In this paper we present a system which utilizes vision and touch for object apprehension. We define apprehension as the determination of the properties of an object and the relationships among these properties. This is in contrast to recognition, which goes a step further in the determination process by attaching a label to the object as a whole. Vision is used to to obtain initial information about the object, including position, two dimensional segmentation, and three dimensional edge analysis. This data is then used to guide an active tactile system in its exploration of the object. The visual and tactile data are combined into a spatial polyhedral representation designed to allow further exploration of the object, as well as high level reasoning about the object and its components.