A Real-Time EMG-Based Fixed-Bandwidth Frequency-Domain Embedded System for Robotic Hand.

Area of Science:

• Biomedical Engineering
• Robotics
• Signal Processing

Background:

• Electromyography (EMG) signals are crucial for volitional control of robotic assistive devices.
• Current Root Mean Square (RMS)-based processing methods face challenges with noise and artifacts, impacting system accuracy.
• Noise and artifacts often exist outside the primary EMG signal frequency bandwidth.

Purpose of the Study:

• To develop a cost-effective embedded system utilizing the Short-Time Fourier Transform (STFT) for EMG-controlled robotic hand applications.
• To identify the optimal myoelectric signal frequency bandwidth for muscle contractions.
• To compare the performance of the STFT embedded system against the conventional RMS embedded system.

Main Methods:

• Healthy volunteers participated to determine the optimal myoelectric signal frequency bandwidth.
• An STFT embedded system was engineered using the STM32 microcontroller unit (MCU).
• Comparative analysis was conducted between the STFT and RMS embedded systems for EMG signal processing.

Main Results:

• The optimal myoelectric signal frequency band for muscle contractions was identified as 60–80 Hz.
• The STFT embedded system demonstrated superior stability in detecting muscle contractions compared to the RMS system.
• The STFT embedded system achieved an average accuracy of 91.55% and did not require onsite calibration, unlike the RMS system.

Conclusions:

• The STFT method provides a robust and accurate approach for processing EMG signals in robotic control.
• The developed embedded system is cost-effective and less complex, offering a novel solution for myoelectric signal processing.
• This advancement has the potential to significantly improve the performance and usability of robotic assistive devices.