Image-based grasp synthesis using Genetic Algorithm

Abstract

This paper presents a new algorithm of grasp-planning synthesis using Genetic Algorithm. The image information of object is taken into consideration for analysis and synthesis. A few fitness functions are defined: some are used to judge whether the individual is an legitimate candidate; the others are used to assess the quality of candidates and rank them for selection and reproduction. An algorithm for synthesizing hand configuration for corresponding grasp is also illustrated, where robot hand's kinematic constraints are taken into account. A few examples of results obtained by developed algorithm are shown in the paper as well.