Koopman-driven grip force prediction through EMG sensing

TL;DR

This paper introduces a novel Koopman operator-based method for real-time grip force prediction from single sEMG sensors, achieving high accuracy and robustness, with potential applications in robotic hand rehabilitation.

Contribution

It presents a new data-driven Koopman approach for accurate, fast grip force estimation and prediction using minimal sEMG sensors, improving rehabilitation device control.

Findings

Achieved approximately 5.5% wMAPE in grip force estimation.

Predicted grip force with 17.9% wMAPE at 0.5-second horizon.

Method executed in about 30 ms, enabling real-time use.

Abstract

Loss of hand function due to conditions like stroke or multiple sclerosis significantly impacts daily activities. Robotic rehabilitation provides tools to restore hand function, while novel methods based on surface electromyography (sEMG) enable the adaptation of the device's force output according to the user's condition, thereby improving rehabilitation outcomes. This study aims to achieve accurate force estimations during medium wrap grasps using a single sEMG sensor pair, thereby addressing the challenge of escalating sensor requirements for precise predictions. We conducted sEMG measurements on 13 subjects at two forearm positions, validating results with a hand dynamometer. We established flexible signal-processing steps, yielding high peak cross-correlations between the processed sEMG signal (representing meaningful muscle activity) and grip force. Influential parameters were subsequently identified through sensitivity analysis. Leveraging a novel data-driven Koopman operator theory-based approach and problem-specific data lifting techniques, we devised a methodology for the estimation and short-term prediction of grip force from processed sEMG signals. A weighted mean absolute percentage error (wMAPE) of approx. 5.5% was achieved for the estimated grip force, whereas predictions with a 0.5-second prediction horizon resulted in a wMAPE of approx. 17.9%. The methodology proved robust regarding precise electrode positioning, as the effect of sensing position on error metrics was non-significant. The algorithm executes exceptionally fast, processing, estimating, and predicting a 0.5-second sEMG signal batch in just approx. 30 ms, facilitating real-time implementation.