An EMG-based robot control scheme robust to time-varying EMG signal features

Area of Science:

• Robotics
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Growing integration of robots in daily life necessitates intuitive control interfaces, particularly for assisting individuals with special needs.
• Conventional human-controlled systems often involve bulky interfaces, limiting user mobility and natural interaction.
• Electromyographic (EMG) signals offer a non-invasive method for capturing human intent through muscle activity.

Purpose of the Study:

• To develop and evaluate a novel human-robot control interface utilizing surface electromyographic (EMG) signals from the human upper limb.
• To enable real-time, intuitive control of an anthropomorphic robot arm in three-dimensional space.
• To ensure the interface is robust to physiological changes such as muscle fatigue and varying contraction levels.

Main Methods:

• Surface EMG electrodes were placed on the user's upper limb to record muscle electrical activity.
• EMG signals were processed to estimate upper limb motion for robot arm control.
• An anthropomorphic robot arm was controlled in real-time using the estimated motion data.
• The system's robustness was tested against time-varying EMG signal characteristics.

Main Results:

• The proposed EMG-based interface successfully enabled real-time control of an anthropomorphic robot arm in 3-D space.
• Motion estimates derived solely from EMG recordings allowed for intuitive user control.
• The interface demonstrated robustness against signal variations caused by muscle fatigue and altered contraction levels.
• Real-time experiments with variable arm motions and speeds validated the system's efficiency.

Conclusions:

• Surface EMG signals provide a viable and natural interface for controlling robot arms, freeing users from cumbersome equipment.
• The developed system offers a robust and efficient solution for human-robot interaction, particularly beneficial for assistive robotics.
• This approach has significant potential for enhancing the independence and quality of life for individuals requiring robotic assistance.