Dextrous Hand-Grasping Preshaping

Abstract

The research conducted in human grasping have emphasized three main phases: the visual location of the target, when the object is analyzed an recognized, the reaching phase of grasping, when the hand preshapes in order to prepare the “shape matching” with the object to grasp, and the third which is the grasping phase. This paper addresses the problem of developing a method that deals with the preshaping phase of grasping and for that we present an indirect approach. For that it is used an image processing algorithm for reconstructing the object that will be grasped and a VR based grasping simulator for validating the stability of the grasp. The result represents a strategy (preshaping and finger trajectories), needed for grasping the object with a real gripper. An evaluation study was conducted investigating the time needed to grasp an object using the presented approach compared with other methods for generating a grasping strategy. The results have shown a significant improvement regarding the time needed for grasping operations on a real gripper.