In this paper, we propose a new stochastic model to compute the grasping force of robot hand. The manipulator arm may be affected by random factors such as weather and smoothness of contact surface in the actual operation process. To balance the external force by choosing a set of appropriate grip force and achieve grip stability and minimum grip force under uncertain factors, a stochastic second order cone programming model is formulated and an appropriate method is proposed to solve it.

grasping force optimization, second-order cone programming, sample average approximation (SAA).