Application of MAG index for optimal grasp planning

Abstract

Grasp planning is one of the interesting problems in robotic object manipulation. Besides, evaluation of grasp configuration has an important role in task execution. Therefore the Multi Aspect Grasp (MAG) performance index is defined for evaluating the grasp quality. In this paper, various methods including Particle Swarm Optimization, Genetic Algorithm, and MATLAB optimization toolbox are used to maximize the MAG index for a given task, in order to find the best grasping point. The results of using these methods are compared, in terms of their speed and accuracy, and it will be concluded that the PSO yield more accurate results.