Fast computation of 4-fingered force-closure grasps from surface points

Abstract

Abshact-This paper addresses the pmblem of computing frictional 4-fingered force-closure grasps of three dimensional objects. The proposed appmach searches for force-closure grasps from a collection of sampled points on the object's surface. Unlike most other works, the appmach b not limited to the objects with a ceertain class of shapes. It can be applied to an object in any shape since only the object's surface poine and corresponding surface normals at the points are n d e d . The efficiency of the appmach arises from a heuristic for search space pruning which is based on ability to efficiently locate regions in three dimensional space rhere fiietion cones inlenect and a randomized lest for checking forcedosure condition. The proposed approach is implemented and preliminary results are pmenled.

Learning hardware agnostic grasps for a universal jamming gripper

“…Several survey papers covering this type of work are available [16]- [18]. Niparnan and Sudsang [19] relaxed the assumption of access to full object geometry, and instead their algorithm searches for forceclosure grasps on sampled points on the object's surface. Huebner et al [20] transformed 3D models to box-based approximation before generating grasp hypotheses.…”

Grasp Planning for a Humanoid Hand

“…As for the works on grasp planning, there are a number of works on contact-level grasp synthesis such as Coelho & Grupen (2004); Niparnan & Sudsang (2004); Ponce et al (1993). As for the grasp planning considering the hand/arm model, Cutkosky (1989) first proposed an expert system to select one of the grasp styles based on grasp quality measures.…”

“…As for the grasp planning based on a random sampling approach, Niparnan & Sudsang (2004) and Borst et al (2003) proposed methods to realize the force closure using the random sampling. Yashima et al (2003) proposed a manipulation planning based on the random sampling.…”

Geometrical approach for grasp synthesis on discretized 3d objects

“…Niparnan and Sudsang [11] generate a number of 4-finger concurrent FC grasps to provide the user with a large set of grasps, so the user can choose an optimum one according to a quality measure appropriate for the particular task. The algorithm is based on the localization of regions in the 3D space where the axes of the friction cones seem to intersect.…”

“…Ding et al [9] propose an algorithm to generate a form-closure grasp with seven frictionless contact points; however, it can be trapped in local minima. Liu et al [10] extend the previous algorithm to find one force-closure grasp with frictional or frictionless contact points, and Niparnan and Sudsang [11] generate several 4-finger concurrent force-closure grasps; their contribution and the comparison with the algorithms presented in this paper will be discussed later. This paper deals with the problem of finding a forceclosure (FC) grasp with frictional or frictionless contact points, and with any number n of contacts, provided that n ≥ 3 for frictional grasps and n ≥ 7 for frictionless grasps.…”