Evaluating the Feasibility of EMG-Based Human-Machine Interfaces for Driving.

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Rehabilitation Technology

Background:

• Upper-limb amputees experience difficulties with daily activities, including driving.
• Electromyography (EMG)-based human-machine interfaces (HMIs) offer a promising assistive technology.
• The efficacy of EMG-HMIs in complex tasks like driving remains under-investigated.

Purpose of the Study:

• To assess the feasibility of EMG-based HMIs for driving.
• To evaluate performance, cognitive workload, usability, and safety of EMG-HMIs.
• To identify areas for improvement in EMG-HMI design for amputees.

Main Methods:

• A driving simulation study was conducted with 19 able-bodied participants.
• Participants used an EMG-based HMI with their dominant and both hands.
• Driving maneuvers, cognitive workload (blink rate, subjective measures), usability (USE questionnaire), and safety were assessed.

Main Results:

• EMG-HMIs resulted in higher lane offset and steering angle but lower steering entropy in some scenarios.
• Cognitive workload was elevated, and usability scores were reduced with EMG-HMIs.
• Safety outcomes were mixed, with better intersection performance but lower overall safety scores.

Conclusions:

• EMG-based HMIs present both challenges and opportunities for high-demand tasks like driving.
• The system shows potential for controlled driving, especially in specific maneuvers.
• Improvements in cognitive workload and usability are crucial for practical application.