Assessment of a Passive Lumbar Exoskeleton in Material Manual Handling Tasks under Laboratory Conditions

Abstract

Manual material handling tasks in industry cause work-related musculoskeletal disorders. Exoskeletons are being introduced to reduce the risk of musculoskeletal injuries. This study investigated the effect of using a passive lumbar exoskeleton in terms of moderate ergonomic risk. Eight participants were monitored by electromyogram (EMG) and motion capture (MoCap) while performing tasks with and without the lumbar exoskeleton. The results showed a significant reduction in the root mean square (VRMS) for all muscles tracked: erector spinae (8%), semitendinosus (14%), gluteus (5%), and quadriceps (10.2%). The classic fatigue parameters showed a significant reduction in the case of the semitendinosus: 1.7% zero-crossing rate, 0.9% mean frequency, and 1.12% median frequency. In addition, the logarithm of the normalized Dimitrov's index showed reductions of 11.5, 8, and 14% in erector spinae, semitendinosus, and gluteus, respectively. The calculation of range of motion in the relevant joints demonstrated significant differences, but in almost all cases, the differences were smaller than 10%. The findings of the study indicate that the passive exoskeleton reduces muscle activity and introduces some changes of strategies for motion. Thus, EMG and MoCap appear to be appropriate measurements for designing an exoskeleton assessment procedure.

Keywords: EMG; exoskeleton; fatigue; lumbar; manual material handling; motion-tracking.

Figure A1

Plot of residues, fitted values minus observed values, over the fitted values for VRMS for each muscle mixed model.

Figure A2

Plot of residues, fitted values minus observed values, over the fitted values for TZC for each muscle mixed model.

Figure A3

Plot of residues, fitted values minus observed values, over the fitted values for FMN for each muscle mixed model.

Figure A4

Plot of residues, fitted values minus observed values, over the fitted values for FMD for each muscle mixed model.

Figure A5

Plot of residues, fitted values minus observed values, over the fitted values for Log(Fimin) for each muscle mixed model.

Figure 1

(Right): Picture of test set up. Structure of the pallet of 16 boxes and participant fully instrumented with exoskeleton, EMG sensors and MoCap inertial sensors. Origin: where boxes on pallet are picked up. Destination: table where user places boxes. (Left): Illustration of setup scheme. Numbers represent the order boxes are picked up.

Figure 2

Assessed lumbar Laevo^TM V2 exoskeleton.

Figure 3

Flow chart of the EMG signals processing. Filtering, segmentation, and parametrization.

Figure 4

Surface EMG signals acquired simultaneously from four muscles. From top to bottom: erector spinae, gluteus medius, quadriceps femoris, semitendinosus. Time window corresponds to last 4 movements of exercise by subject 4 without exoskeleton. Vertical marks indicate beginning (green) and end (red) of myoelectrical activation, before visual check carried out to catch the boxes.

Figure 5

Marginal mean curves with the standard error bars of VRMS parameter throughout the 16 boxes for all four muscles: (a) Erector Spinae, (b) Gluteus, (c) Quadriceps, and (d) Semitendinosus.

Figure 6

Marginal mean curves with the standard error bars of TZC parameter throughout the 16 boxes for all four muscles: (a) Erector Spinae, (b) Gluteus, (c) Quadriceps, and (d) Semitendinosus.

Figure 7

Marginal mean curves with the standard error bars of FMD parameter throughout the 16 boxes for all four muscles: (a) Erector Spinae, (b) Gluteus, (c) Quadriceps, and (d) Semitendinosus.

Figure 8

Marginal mean curves with the standard error bars of FMN parameter throughout the 16 boxes for all four muscles: (a) Erector Spinae, (b) Gluteus, (c) Quadriceps, and (d) Semitendinosus.

Figure 9

Marginal mean curves with the standard error bars of Log(FImin) parameter throughout the 16 boxes for all four muscles: (a) Erector Spinae, (b) Gluteus, (c) Quadriceps, and (d) Semitendinosus.