Genetic Algorithms for Stable Robot Grasping

Abstract

The goal of this paper is to propose an optimal positioning of robotic hand fingers, for a generic 3D object, to ensure stable grasping for an industrial pick and place process. The proposed strategy for optimal positioning based on genetic algorithms is presented. The grasping configuration is determined under several criteria that ensure the object stability. One criterion is based on Static Stability Margin (SSM) that takes into consideration the position of object center of mass in the polygon of support and a second criterion based on the Force Closure Grasping (FCG) taking into consideration the contact forces applied by the robotic hand fingertips. The optimal positioning algorithm results are presented to validate our proposal for a different kind of products.