The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge

Research Paper #769

Title:Planning Dextrous Hand Precision Grasps from Range Data, Using Preshaping and Finger Trajectories
Authors:Wren,DO; Fisher,RB
Date:Aug 1995
Presented:Presented at the Third International Symposium on Intelligent Robotic Systems, Pisa, 1995
Abstract:Inspired by human grasping behaviour, use of hand preshapes has for some time been recognised as a useful way of reducing the complexity of planning dextrous hand gasps. It is common for preshapes to be chosen and fitted to the object using coarse object characteristics. In this paper we use range data to provide the more detailed knowledge of surface geometry required when planning precision grasps. We define sets of preshapes with associated modes of fingertip closure, and give criteria for the choice and positioning of the preshapes based on intersecting fingertip trajectories with the graspable surfaces of the object. Grasps are planned in simulation using real range data and an anthropomorphic hand model. Our results show that complex objects can be grasped using relatively simple, fast algorithms when the hand movements are constrained in this way.

[Search These Pages] [DAI Home Page] [Comment]