EMG Based Motion Intension Recognition and Force Estimation
Wang, Xiaoyu (2020-01-07)
EMG Based Motion Intension Recognition and Force Estimation
(07.01.2020)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe202003259227
https://urn.fi/URN:NBN:fi-fe202003259227
Tiivistelmä
Hands are the most dexterous organs of human beings. People communicate with each other with various hand gestures. However, it is difficult for stroke survivors and transradial or forearm amputees to perform gestures in daily lives. As a result, it is of great significance for hand gestures related researches in medical rehabilitation area.
EMG based hand gesture recognition has been widely used in the related area because of the advantages of non-invasive and less interference of EMG signal. According to the data acquisition methods, EMG can be divided into needle EMG and surface EMG. Needle EMG is invasive and is usually used for pathological analysis. Thus we only focus on the surface EMG signal. It is collected by conventional electrodes and high density electrodes. Conventional electrodes have been widely used in commercial use. And high density EMG is regarded as a new and promising control signal in human machine system because it provides more spatial and micro features. Collected EMG signal is used for rehabilitation and prosthesis control by pattern recognition or proportional control.
This study mainly focuses on EMG based gesture recognition and force estimation in rehabilitation and prosthesis control. The feasibility of upper arm EMG based wrist and finger gesture recognition for transradial and forearm amputees has been proved in this study and it makes it more convenient for their daily communication. Furthermore, the micro method of spike train based force estimation contributes to the EMG based fine prosthesis control and is beneficial for the stroke survivors and amputees. It can be concluded that the results in this study are meaningful in the related research area and highly promotes the development of the area of medical rehabilitation and prosthesis control.
EMG based hand gesture recognition has been widely used in the related area because of the advantages of non-invasive and less interference of EMG signal. According to the data acquisition methods, EMG can be divided into needle EMG and surface EMG. Needle EMG is invasive and is usually used for pathological analysis. Thus we only focus on the surface EMG signal. It is collected by conventional electrodes and high density electrodes. Conventional electrodes have been widely used in commercial use. And high density EMG is regarded as a new and promising control signal in human machine system because it provides more spatial and micro features. Collected EMG signal is used for rehabilitation and prosthesis control by pattern recognition or proportional control.
This study mainly focuses on EMG based gesture recognition and force estimation in rehabilitation and prosthesis control. The feasibility of upper arm EMG based wrist and finger gesture recognition for transradial and forearm amputees has been proved in this study and it makes it more convenient for their daily communication. Furthermore, the micro method of spike train based force estimation contributes to the EMG based fine prosthesis control and is beneficial for the stroke survivors and amputees. It can be concluded that the results in this study are meaningful in the related research area and highly promotes the development of the area of medical rehabilitation and prosthesis control.