International Journal of applied mathematics and computer science

online read us now

Paper details

Number 2 - June 2016
Volume 26 - 2016

Dynamic optimal grasping of a circular object with gravity using robotic soft-fingertips

Rodolfo García-Rodríguez, Victor Segovia-Palacios, Vicente Parra-Vega, Marco Villalva-Lucio

Abstract
Object manipulation usually requires dexterity, encoded as the ability to roll, which is very difficult to achieve with robotic hands based on point contact models (subject to holonomic constraints). As an alternative for dexterous manipulation, deformable contact with hemispherical shape fingertips has been proposed to yield naturally a rolling constraint. It entails dexterity at the expense of dealing with normal and tangential forces, as well as more elaborated models and control schemes. Furthermore, the essential feature of the quality of grasp can be addressed with this type of robot hands, but it has been overlooked for deformable contact. In this paper, a passivity-based controller that considers an optimal grasping measure is proposed for robotic hands with hemispherical deformable fingertips, to manipulate circular dynamic objects. Optimal grasping that minimizes the contact wrenches is achieved through fingertip rolling until normal forces pass through the center of mass of the object, aligning the relative angle between these normal forces. The case of a circular object is developed in detail, though our proposal can be extended to objects with an arbitrary shape that admit a local decomposition by a circular curvature. Simulation and experimental results show convergence under various conditions, wherein rolling and tangent forces become instrumental to achieve such a quality of grasp.

Keywords
dynamic grasping, soft-fingertips, circular object

DOI
10.1515/amcs-2016-0022