A multi-dimensional surface electromyographic signal prosthetic hand control method based on principal component analysis

Abstract

The invention discloses a multi-dimensional surface electromyography signal prosthetic hand control method based on principal component analysis, which comprises the following steps: firstly, an armband equipped with a 24-channel array electromyography sensor is worn on the forearm of a subject, and five finger joints are The attitude sensors were worn on the distal phalanx of the subject's thumb and the middle phalanx of the rest of the fingers; the subject performed independent bending and stretching training of five fingers, and at the same time collected myoelectric sensor array data and finger joint attitude sensor data; using principal components The analysis method decouples the myoelectric sensing data to form a finger movement training set; after the training is completed, the sensor worn on the finger is removed; the neural network method is used to perform data fitting on the above finger movement training set to construct continuous finger movement Predictive model; use the continuous finger motion model to predict the current bending angle of the finger. The invention can overcome the incoherence of discrete action mode classification, and finally achieve smoother control of the prosthetic hand.

Description

technical field [0001] The invention relates to a control method for a manipulator, in particular to a method for controlling a prosthetic hand based on a principal component analysis method with multi-dimensional surface electromyography signals. Background technique [0002] Biomechatronics dexterously operated prosthesis is an intelligent interactive device that can work with the environment, humans and other robots. It recognizes the operator's action intentions by collecting bioelectrical signals from the human body. Research on artificial limbs can drive technological innovation in the field of functional reconstruction and rehabilitation engineering for the disabled, and extend and develop the scientific connotation of equipment manufacturing. Its scientific and technological achievements can be radiated and applied to high-end medical equipment, mechanical and electrical integration intelligent robots, hazardous environment exploration and disaster rescue equipment, ...

AI Technical Summary

Problems solved by technology

[0004] Due to the different degrees of forearm muscle development and operating habits of different subjects, it is often difficult to extract universal motion rules for all individuals.

[0005] Human finger movements are very complex. Existing research focuses on the recognition of isolated gestures, and rarely recognizes continuous movements of gestures.

This technology uses the principal component analysis method to decouple the complex muscle activities of the hand, and can analyze the continuous activities of each finger. At present, there is no literature on the continuous motion estimation of fingers.

Images