Abstract:

Objective To establish a multi index fusion hand grip fatigue prediction model to evaluate the power-assisted effect of the glove exoskeleton prototype for extravehicular clothing.

Methods BP neural network algorithm was used to establish a hand fatigue prediction model. The related factors of hand fatigue were determined with isometric grasping fatigue experiment, and the input variables of BP neural network were determined as cylinder diameter, grasping force, grasping duration and root mean square of electromyography. The fatigue data corresponding to variables of each group were obtained through experiments and subjective fatigue measurement scales, and a fatigue evaluation model based on multi-source fusion of BP neural network algorithm was established. The relationship model between fatigue and assistance effect was established, and the assistance effect of the exoskeleton prototype was evaluated through the degree of fatigue relief.

Results The correlation coefficient was 0.974 between the predicted results of the model and the target value. Moreover, it effectively predicted the assistance effect of different prototypes.

Conclusion The BP neural network model established by combining the grasping strength, grasping object parameters and human electromyography can predict hand fatigue, which can be used to evaluate the assistance effect of glove exoskeleton and other hand aids.

Key words: glove exoskeleton, extravehicular activity, hand, fatigue, neural network, evaluation