Feasibility Study of Upper Limb Control Method Based on Electromyography-Angle Relation

Abstract

The paper describes the method of predicting the angular position of the human upper limb using EMG signals. A neural network with fuzzy logic was used for this purpose. The main goal of the work, namely, to demonstrate that a neural network with fuzzy logic is a useful tool for predicting motion based on EMG signals, has been completed. Two EMG signals from those muscles of the human arm that show the greatest activity during the load lifting are used. When determining the driving torques, the differences between the intended and the actual angular position are taken into account, and a simplified dynamics model was used for the calculations. In order to validate the method, the actual and predicted angles are compared and the differences between the moments determined on the basis of anticipated angular positions and the moments provided by the opensim simulator using real angular positions are examined.