Qualitative Framework for Vision-based Closure Grasps

Shyamanta M Hazarika

ABSTRACT: Vision based grasping formalisms require complete quantitative knowledge and are often susceptible to slight errors in shape and position of the object. Driven by the fact that everyday spatial reasoning is through qualitative abstractions, the focus of this paper is towards development of a qualitative framework for synthesis of closure grasps. The spatial representation language RCC-8, often referred to as Region Connection Calculus and its spatio-temporal extension, ST0, a tractable fragment of Propositional Spatio-temporal Logic is used as the knowledge representation formalism. Qualitative grasping schemes within this framework are explored and set of algorithms for synthesizing closure grasps is presented.

KEYWORDS: Grasping, Closure Grasp, Qualitative Kinematics, Qualitative Spatial Reasoning

full paper (pdf)