Related Experiment Video
Updated: Sep 22, 2025
06:58
A Structured Rehabilitation Protocol for Improved Multifunctional Prosthetic Control: A Case Study
Published on: November 6, 2015
9.6K
Real-Time Control of Intelligent Prosthetic Hand Based on the Improved TCN
Xiaoguang Liu1,2, Jiawei Wang1,2, Tingwen Han1,2
1College of Electronic and Information Engineering, Hebei University, Baoding, Hebei, China.
Applied Bionics and Biomechanics
|May 24, 2022
Summary
This study introduces a new multichannel fusion scheme (MSFS) and an improved Temporal Convolutional Network (TCN) for enhanced surface electromyography (sEMG) signal processing. This advancement significantly improves the accuracy of controlling intelligent prosthetic hands, achieving a 93.69% success rate.
Area of Science:
- Robotics
- Biomedical Engineering
- Artificial Intelligence
Background:
- Intelligent prosthetic hands are crucial for remote task completion and rehabilitation.
- Surface electromyography (sEMG) is a key technology for accurate prosthetic hand control.
- Existing sEMG methods face limitations in accuracy and channel extension.
Purpose of the Study:
- To propose a novel multichannel fusion scheme (MSFS) for extending sEMG virtual channels.
- To enhance the Temporal Convolutional Network (TCN) for improved gesture recognition performance.
- To validate the efficacy of the proposed method in real-time prosthetic hand control.
Main Methods:
- Development of a new multichannel fusion scheme (MSFS) for sEMG signal processing.
- Improvement of a Temporal Convolutional Network (TCN) architecture for enhanced deep learning capabilities.
- Real-time data collection using a Myo armband for sEMG signal acquisition.
- Real-time control of an intelligent prosthetic hand to validate the proposed method.
Main Results:
- The proposed MSFS effectively extends virtual sEMG channels, enhancing gesture recognition.
- The improved TCN demonstrates superior performance in processing complex sEMG data.
- Experimental validation shows a significant improvement in the accuracy of intelligent prosthetic hand control.
- The achieved accuracy rate for prosthetic hand control reached 93.69%.
Conclusions:
- The developed multichannel fusion scheme and enhanced TCN offer a robust solution for sEMG-based prosthetic control.
- The proposed method significantly improves the accuracy and reliability of intelligent prosthetic hands.
- This research contributes to advancing human-computer interaction in the field of intelligent robotics and prosthetics.
