Objective:

In the design of a robotic hand prosthesis, a variety of problems arise for which there are multiple solutions proposed in the literature. One of them is the selection of a strategy for the automatic control of the movement of the hand. Within the functionalities of a prosthesis, the power grip functionality is one of the most important. Given the wide variety of proposals in terms of control techniques for power grip, the designer is faced with the dilemma of which of them is the most suitable for his particular design. Although there are metrics to quantify the quality of the grip, those that have been proposed address various aspects independently, which does not allow making a decision about the best option for the particular case. In this work, a method is proposed to calculate a composite indicator that allows the evaluation of the performance of control techniques in the power grip based on the following individual metrics that evaluate different aspects in the execution of the power grip: finger strength, grip strength, grip efficiency, grip cycle time, slip resistance, following error, and transient response overshoot. Thanks to the method proposed in this work, the question is answered: of several control options to be tested, which one offers better functional performance?

Methodology:

A set of metrics are adopted with which to obtain quantitative data related to the quality of the power grip and to the performance of the control target to govern the prosthesis; subsequently, selected individual metrics are calculated for each of the grip control techniques to be evaluated on a virtual environment, consisting of a robotic hand and an object to be grasped; then two composite indicators are constructed to obtain a quantitative assessment of the quality of the grip based on a statistical analysis, and the results are contrasted against the individual metrics used.

Results:

A method is proposed for the construction of a composite indicator, which allows the evaluation of the performance of control techniques, in power grips in robotic hands. By implementing this method, the best performance values were found in hybrid controllers.

Conclusions:

In this work, a method has been proposed to facilitate the decision-making of the designer regarding the most adequate control technique, among several available, to achieve the power grip with a specific prosthesis. The method seeks to build a composite indicator that groups together various metrics to evaluate particular grip functionality, and also to quantify the achievement of following instructions, facilitating decision-making about the incidence of the control technique in the achievement of the final objective.

Financing:

This work was supported by University of Cauca under the Master of Science in Automation program.