Abstract: When grasping an object with a multi-fingered robot hand, contact points on the object should be arranged deliberately to achieve maximum resistance to external disturbances. Taking the maximum external wrench in all directions a grasp can balance as a criterion, a planning algorithm is proposed to calculate the initial positions of the fingers for polygonal objects grasping, satisfying the force closure condition. After careful analysis of the features of the primitive wrench set and both the geometrical and the physical significances of the grasping planning criterion in wrench space, a linear programming can be designed to calculate the optimal direction to improve grasp performance based on the given initial grasp arrangement. Starting at the initial arrangement, an iterative algorithm is proposed to continuously improve grasp performance towards the optimal direction and finally converge to the optimal grasp arrangement. Numerical examples show that the initial arrangement with relative good performance can be obtained by the initial arrangement planning algorithm. Meanwhile, the optimal grasping planning algorithm improves the performance towards the optimal direction at every iterative step. Thus, the planning algorithm runs in less iteration with higher efficiency. The final solution reaches or approaches global optimum of the grasping arrangement.

Key words: Force closure, Multi-fingered hand, Optimal grasp planning, Wrench space