A multi-threshold-based force regulation policy for prosthetic hand preventing slippage

Abstract

The slippage occurs between the fingers and object will result in grasp failure that badly affects the prosthetic hand's operation. In this paper, a new slippage prevention strategy is proposed to improve the prosthetic hand's grasp stability. The proposed method, which comprises a slipping process model, a multi-threshold detection scheme, and a force regulation policy, can have a minimal loading force intervention after stabilization. First, the slipping process is modeled with special considerations on the nonlinear and noise characteristics of the system. Then, the multiple thresholds are selected for slippage detection to reduce the influence of loading force variation and system nonlinearity. Lastly, a loading force regulation policy consisting of a serials of constant, slight force increments is applied to suppress the detected slippage and minimize the force growth. One merit of our method is that the parameters selected in the algorithm largely depend on the system characteristics of the hand system, but not the grasped object. Experimental results show that the proposed method has better detection accuracy on slippage detection and minimal loading force intervention after slippage suppression.