EMG-based power assist control system

• Main Author: Xing, Yi Tong.
• Other Authors: Low Kay Soon
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Medical electronics
• Online Access: http://hdl.handle.net/10356/39714
• Institution: Nanyang Technological University

Electromyography (EMG) is an electrical signal generated in the muscles when the muscles contract. An EMG power assist system is a system which uses an EMG interface to interpret the intentions of the human from the EMG signals and thus control an actuator which helps perform the task, allowing the human strength to be increased. Raw EMG signals have to be first processed to ensure a certain quality to the EMG signal before the features of the signal can be extracted and used for motion discrimination. In this project, 2 degrees of motions of the arms were considered. Experiments were then conducted to study the relationship between joint angles and the EMG signals. A method to recognize the EMG waveform when carrying various arm motions was also introduced. Finally, a real(time system utilizing an Artificial Neural Network to predict joint angle was also implemented and tested on a human subject. This report summarizes the processing techniques used to obtain a quality EMG waveform as well as how EMG signals could be used to identify arm motions and joint angles.