Human-Inspired Force Compliant Grasping Primitives

We address the problem of grasping everyday ob- jects that are small relative to an anthropomorphic hand, such as pens, screwdrivers, cellphones, and hammers from their nat- ural poses on a support surface, e.g., a table top. In such con- ditions, state of the art grasp generation techniques fail to pro- vide robust, achievable solutions due to either ignoring or try- ing to avoid contact with the support surface. In contrast, when people grasp small objects, they often make use of substan- tial contact with the support surface. In this paper we give re- sults of human subjects grasping studies which show the ex- tent and characteristics of environment contact under differ- ent task conditions. We develop a simple closed-loop hybrid grasping controller that mimics this interactive, contact-rich strategy by a position-force, pre-grasp and landing strategy for finger placement. The approach uses a compliant control of the hand during the grasp and release of objects in order to preserve safety. We conducted extensive robotic grasping ex- periments on a variety of small objects with similar shape and size. The results demonstrate that our approach is robust to lo- calization uncertainties and applies to many everyday objects.