Estimation of User-Applied Isometric Force/Torque Using Upper Extremity Force Myography

Abstract

Hand force estimation is critical for applications that involve physical human-machine interactions for force monitoring and machine control. Force Myography (FMG) is a potential technique to be used for estimating hand force/torque. The FMG signals reflect the volumetric changes in the arm muscles due to muscle contraction or expansion. This paper investigates the feasibility of employing force-sensing resistors (FSRs) worn on the arm to measure the FMG signals for isometric force/torque estimation. Nine participants were recruited in this study and were asked to exert isometric force along three perpendicular axes, torque about the same three axes, and force and torque simultaneously. During the tests, the isometric force and torque were measured using a 6-degree-of-freedom (DoF) (i.e., force in three axes and torque around the same axes) load cell for ground truth labels whereas the FMG signals were recorded using a total number of 60 FSRs, which were embedded into four bands worn on the different locations of the arm. A two-stage regression strategy was employed to enhance the performance of the FMG bands, where three regression algorithms including general regression neural network (GRNN), support vector regression (SVR), and random forest regression (RF) models were employed, respectively, in the first stage and GRNN was used in the second stage. Two cases were considered to explore the performance of the FMG bands in estimating: (1) 3-DoF force and 3-DoF torque at once and (2) 6-DoF force and torque. In addition, the impact of sensor placement and the spatial coverage of FMG measurements were studied. This preliminary investigation demonstrates promising potential of FMG to estimate multi-DoF isometric force/torque. Specifically, R 2 accuracies of 0.83 for the 3-DoF force, 0.84 for 3-DoF torque, and 0.77 for the combination of force and torque (6-DoF) regressions were obtained using the four bands on the arm in cross-trial evaluation.

A Review of EMG-, FMG-, and EIT-Based Biosensors and Relevant Human–Machine Interactivities and Biomedical Applications

“…Sakr et al used FSR 402 to estimate hand force using 16 channels at a 10 Hz sampling frequency. Finally, they obtained the R 2 accuracies of 0.83 for the 3-DoF force, 0.84 for 3-DoF torque, and 0.77 for the combination of force and torque (6-DoF) in cross-trial evaluation [ 42 ]. Except for the most widely used FSR sensor, the sampling frequency of other sensors is different, but most of them choose a lower sampling frequency to match the human body movement.…”

“…Because FMG has the characteristics of intuitively reflecting muscle activity, it can reflect the magnitude of force by detecting the intensity of muscle activity to a certain extent. Sakr et al use a total of 60 FSRs embedded in four bands, estimating in two cases: (1) 3-DoF force and 3-DoF torque at once and (2) 6-DoF force and torque [ 42 ]. The results showed that FMG achieves a good performance in multiple-DoF force/torque estimation.…”

Biosignal-Based Human–Machine Interfaces for Assistance and Rehabilitation: A Survey

“…Several studies show the potential of FMG as an alternative control strategy for HMI applications, particularly for upper-limb prosthetic control [9]. In this context, Sakr et al demonstrated the feasibility of estimating hand isometric force/torque from FMG signals recorded via 60 FSRs, which were embedded into four bands and placed in different locations around the arm [149]. Again, Sakr et al showed the possibility of predicting force in dynamic conditions by using FSRs worn around the arm [150].…”

Unsupervised, Semi-Supervised Interactive Force Estimations During pHRI via Generated Synthetic Force Myography Signals

“…Estimating isometric hand force from wrist, forearm, and upper arm FMG signals were found promising in [13]. Although a loadcell or a force/torque (FT) sensor can read forces precisely, it is bulky, requires complex signal processing units and is difficult to move freely when used as wearable on a human body.…”