Force Myography for Motion Intention Detection Based on 3D-Printed Piezoelectric Sensors

Recognizing human movement intention is key to control wearable robotics for assistance such as in the elderly and rehabilitation. Piezoelectric sensors, in particular highly sensitive ferroelectrets (FEs), qualify for the assessment of subtle muscle movements using force myography (FMG). In contrast to electromyography (EMG), which measures electrical muscle activity, FMG measures the mechanical muscle response to overcome issues in electrode placement and skin preparation. We employ 3D-printed polypropylene FE sensor patches for FMG adjacent to EMG electrodes as a reference at the lower limb of six young unimpaired participants performing sit-to-stand movements. The activity onsets of FMG and EMG are systematically compared at the quadriceps femoris and visually verified using a high-speed camera. The average time difference (35±50 ms) aligns with the known electromechanical muscle delay. Notably, our results demonstrate that FMG can provide an early indicator for the activity onset significantly before the EMG (on average 263
$\pm$
90 ms earlier) in this scenario due to the possibility to measure passive muscle deformation. In addition, the signal-to-noise ratio (SNR) of the FE signal (53 dB) exceeds that of the EMG (39 dB) during the maximum muscle activity, while the average variability is slightly larger for FE (22%) compared to EMG (17%). In conclusion, our customizable FE sensors prove to be suitable for motion intention detection and promise an advance in wearable assistive devices, such as active prostheses and exoskeletons, to support people in need.