An EEG/EMG/EOG-Based Multimodal Human-Machine Interface to Real-Time Control of a Soft Robot Hand

Area of Science:

• Neuroscience
• Robotics
• Rehabilitation Engineering

Background:

• Brain-computer interface (BCI) technology offers promise for motor rehabilitation in stroke survivors by leveraging neural plasticity.
• Current BCI systems face challenges in providing diverse control commands for natural, multi-task soft robot operation.

Purpose of the Study:

• To develop a novel multimodal human-machine interface (mHMI) for enhanced control of soft robots.
• To assess the feasibility and user acceptance of an affordable wearable soft robot for assisted hand movements.
• To investigate the combined efficacy of electrooculography (EOG), electroencephalography (EEG), and electromyogram (EMG) signals.

Main Methods:

• Developed an mHMI integrating EOG, EEG, and EMG signals for generating multiple control instructions.
• Six healthy subjects performed motor imagery, eye movements, and hand gestures to control a soft robot.
• Evaluated the number of control instructions, classification accuracy, and information transfer rate.

Main Results:

• The mHMI generated a significantly greater number of control instructions compared to individual signal modalities.
• Achieved an average classification accuracy of 93.83% and an information transfer rate of 47.41 bits/min.
• Demonstrated a control speed equivalent to 17 actions per minute.

Conclusions:

• The developed mHMI system offers a user-friendly and effective solution for real-time soft robot control.
• This technology has the potential to aid both healthy and disabled individuals in performing basic hand movements.
• The findings support the advancement of BCI applications in assistive robotics and motor rehabilitation.