A parallel robots framework to study precision grasping and dexterous manipulation
Júlia Borràs, Aaron M. Dollar
- Year
- 2013
- Citations
- 24
Abstract
Dexterous, within-hand manipulation, in which an object generally held in the fingertips is manipulated by the fingers, shares many similarities to parallel robot configurations. This paper shows how to apply a mathematical framework commonly used for parallel robots to study the kinetostatic properties of hands manipulating objects using precision grasps, considering compliance and underactuation in the joints, without requiring the use of the grasp matrix. The proposed framework is suitable for any hand, but we focus on underactuated hands. We show how the natural redundancy present in fully-actuated hands can be eliminated using underactuation, leading to simplified non-redundant systems that are easier to control. We primarily focus our efforts on introducing and describing the theoretical framework, and follow this with an example application using a three-fingered underactuated hand. For this example, we define the feasible workspace as the subspace of the kinematic workspace for which the hand can accomplish a grasp, and we study how the compliance, rest angles, and joint coupling in the fingers can be designed to increase the size of this feasible workspace.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002