Optimizing Calibration Procedure to Train a Regression-Based Prediction Model of Actively Generated Lumbar Muscle Moments for Exoskeleton Control

The risk of low-back pain in manual material handling could potentially be reduced by back-support exoskeletons. Preferably, the level of exoskeleton support relates to the required muscular effort, and therefore should be proportional to the moment generated by trunk muscle activities. To this end, a regression-based prediction model of this moment could be implemented in exoskeleton control. Such a model must be calibrated to each user according to subject-specific musculoskeletal properties and lifting technique variability through several calibration tasks. Given that an extensive calibration limits the practical feasibility of implementing this approach in the workspace, we aimed to optimize the calibration for obtaining appropriate predictive accuracy during work-related tasks, i.e., symmetric lifting from the ground, box stacking, lifting from a shelf, and pulling/pushing. The root-mean-square error (RMSE) of prediction for the extensive calibration was 21.9 nm (9% of peak moment) and increased up to 35.0 nm for limited calibrations. The results suggest that a set of three optimally selected calibration trials suffice to approach the extensive calibration accuracy. An optimal calibration set should cover each extreme of the relevant lifting characteristics, i.e., mass lifted, lifting technique, and lifting velocity. The RMSEs for the optimal calibration sets were below 24.8 nm (10% of peak moment), and not substantially different than that of the extensive calibration.

[1]  Frank Krause,et al.  Exoskeletons for industrial application and their potential effects on physical work load , 2016, Ergonomics.

[2]  D Chaffin,et al.  High-pass filtering to remove electrocardiographic interference from torso EMG recordings. , 1993, Clinical biomechanics.

[3]  M. Frings-Dresen,et al.  The effect of lifting during work on low back pain: a health impact assessment based on a meta-analysis , 2014, Occupational and Environmental Medicine.

[4]  J H van Dieën,et al.  Are recruitment patterns of the trunk musculature compatible with a synergy based on the maximization of endurance? , 1997, Journal of biomechanics.

[5]  Mantian Li,et al.  Towards Online Estimation of Human Joint Muscular Torque with a Lower Limb Exoskeleton Robot , 2018, Applied Sciences.

[6]  Gert S. Faber,et al.  How to lift a box that is too large to fit between the knees , 2010, Ergonomics.

[7]  Darwin G. Caldwell,et al.  Evaluation of an acceleration-based assistive strategy to control a back-support exoskeleton for manual material handling , 2021, Wearable Technologies.

[8]  J H van Dieën,et al.  Effects of antagonistic co-contraction on differences between electromyography based and optimization based estimates of spinal forces , 2005, Ergonomics.

[9]  W S Marras,et al.  A Three-Dimensional Motion Model of Loads on the Lumbar Spine: I. Model Structure , 1991, Human factors.

[10]  Idsart Kingma,et al.  Ltd Registered in England and Wales Registered Number : 1072954 Registered office : , 2011 .

[11]  Axel S. Koopman,et al.  The effect of control strategies for an active back-support exoskeleton on spine loading and kinematics during lifting. , 2019, Journal of biomechanics.

[12]  I. Kingma,et al.  Validation of a full body 3-D dynamic linked segment model , 1996 .

[13]  R. Norman,et al.  Comparison of muscle forces and joint load from an optimization and EMG assisted lumbar spine model: towards development of a hybrid approach. , 1995, Journal of biomechanics.

[14]  Günter Hommel,et al.  A Human--Exoskeleton Interface Utilizing Electromyography , 2008, IEEE Transactions on Robotics.

[15]  M. Adams,et al.  Influence of lumbar and hip mobility on the bending stresses acting on the lumbar spine. , 1993, Clinical biomechanics.

[16]  Michiel P de Looze,et al.  Selecting the appropriate input variables in a regression approach to estimate actively generated muscle moments around L5/S1 for exoskeleton control. , 2020, Journal of biomechanics.

[17]  R. Norman,et al.  Mechanically corrected EMG for the continuous estimation of erector spinae muscle loading during repetitive lifting , 2004, European Journal of Applied Physiology and Occupational Physiology.

[18]  J H van Dieën,et al.  Validation of a wearable system for 3D ambulatory L5/S1 moment assessment during manual lifting using instrumented shoes and an inertial sensor suit. , 2020, Journal of biomechanics.

[19]  Rob J Hyndman,et al.  Another look at measures of forecast accuracy , 2006 .

[20]  B Schibye,et al.  Mechanical load on the low back and shoulders during pushing and pulling of two-wheeled waste containers compared with lifting and carrying of bags and bins. , 2001, Clinical biomechanics.

[21]  Scott Doyle,et al.  Role of training data variability on classifier performance and generalizability , 2018, Medical Imaging.

[22]  S. McGill Electromyographic activity of the abdominal and low back musculature during the generation of isometric and dynamic axial trunk torque: Implications for lumbar mechanics , 1991, Journal of orthopaedic research : official publication of the Orthopaedic Research Society.

[23]  R. Norman,et al.  A comparison of peak vs cumulative physical work exposure risk factors for the reporting of low back pain in the automotive industry. , 1998, Clinical biomechanics.

[24]  Karen J. Reynolds,et al.  Removing power line noise from recorded EMG , 2001, 2001 Conference Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[25]  Luigi Monica,et al.  Back-Support Exoskeletons for Occupational Use: An Overview of Technological Advances and Trends , 2019, IISE Transactions on Occupational Ergonomics and Human Factors.

[26]  C Larivière,et al.  Comparative ability of EMG, optimization, and hybrid modelling approaches to predict trunk muscle forces and lumbar spine loading during dynamic sagittal plane lifting. , 2001, Clinical biomechanics.