A simplified model of a multi-jointed mechanical finger calibrated with experimental data by vision system

Abstract

The development of suitable models for mechanical fingers, whether they are part of prosthetic device or of a robotic hand, is a powerful tool to predict the behaviour of their components since the early stages of design, especially for underactuated mechanisms. Experimental data can improve the reliability of such models and promote their application to build proper control strategies especially for prosthetic hands.

Here, we have developed a multi-jointed model of a mechanical finger. The finger is part of the Federica hand: an underactuated mechanical hand that was conceived for prosthetic purpose. The model accounts for friction phenomena in the finger and it is tuned with experimental data acquired through a digital image correlation device. The model allowed us to write kinematics relations of the phalanges and evaluate finger configurations in relation to the closure velocity. Moreover, it was possible to estimate the tendon force and the work analysis occurring during the closure tasks, both in free mode and in presence of objects.