A method for continuous estimation of human joint angles based on surface EMG signals

Abstract

The invention belongs to the field of human-computer interaction, and discloses a method for continuously estimating the joint angle of a human body based on surface electromyographic signals. The method comprises the following steps: (a) selecting the joint part of the object to be predicted, and collecting the myoelectric signal and the joint angle signal of the joint part; (b) preprocessing the collected myoelectric signal and the joint angle signal respectively; (c ) using the preprocessed myoelectric signal and joint angle signal, using the sparse pseudo-input Gaussian process regression algorithm to construct a prediction model about the myoelectric signal and joint angle signal; (d) collecting the real-time myoelectric signal input prediction model of the object to be predicted In order to obtain the required joint angle signal. Through the present invention, the movement of surface electromyographic signals mapped to joint angles is realized in a continuous manner, which ensures real-time performance of continuous estimation, improves prediction accuracy, and reduces computational complexity.

Description

technical field [0001] The invention belongs to the field of human-computer interaction, and more specifically relates to a method for continuously estimating the joint angle of a human body based on surface electromyographic signals. Background technique [0002] "Human-computer natural interaction" as an important component module has attracted people's attention. Human surface electromyography (sEMG) can be used as a human body because it is easy to collect, contains information about human motion intentions, and its signal generation is ahead of motion. The interface of computer interaction is used to realize the natural interaction between human and computer. Current commercial devices that use surface electromyography signals as human-machine interfaces are usually limited in range of motion, sensory feedback, and limited functions in intuitive control. Although the mechanical structure of the device has multiple degrees of freedom and can complete flexible actions, it...

AI Technical Summary

Problems solved by technology

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a method for continuously estimating joint angles based on surface electromyographic signals, by using a sparse pseudo-input Gaussian process regression algorithm to construct a mapping model between electromyographic signals and human joint angles, Realize the movement of the surface electromyographic signal mapped to the joint angle in a continuous manner to ensure the real-time performance of the continuous estimation method, thus solving the technical problems of low prediction accuracy, poor real-time performance and complex calculation of the continuous estimation method

Images