A Comparison of Myoelectric Control Modes for an Assistive Robotic Virtual Platform

Area of Science:

• Robotics
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Assistive robotics and prosthetics require intuitive control methods.
• Virtual environments offer a safe and accessible platform for training.
• Surface electromyography (sEMG) shows potential for non-invasive neural control.

Purpose of the Study:

• To prove the feasibility of virtual environments controlled by sEMG for training individuals with upper limb impairments.
• To evaluate simple control algorithms against complex machine learning approaches for prosthesis control.
• To explore the benefits of intelligent assistance in robotic grasping tasks.

Main Methods:

• A virtual kitchen environment was created for object grasping and storage tasks.
• A virtual robot arm was controlled using different myoelectric control modes based on sEMG signals.
• Participants performed tasks using various control strategies, including simple algorithms and intelligent assistance.

Main Results:

• High performance was achieved across all participants in the virtual environment.
• Participants reported positive experiences and similar opinions on the tested control modes.
• Simple control algorithms demonstrated robustness and natural interaction.

Conclusions:

• sEMG-controlled virtual environments are feasible for training users of prosthetics and assistive devices.
• Simple control algorithms provide a more natural and robust interaction for assistive robotics compared to complex ML approaches.
• Further research into intelligent assistance can enhance grasping activities in assistive robotics.