Reconstruction of occluded pelvis markers during marker-based motion capture with industrial exoskeletons

Abstract

Industrial back support exoskeletons are a promising solution to alleviate lumbar musculoskeletal strain. Due to the complexity of spinal loading, evaluation of EMG data alone has been considered insufficient to assess their support effects, and complementary kinematic and dynamic data are required. However, the acquisition of marker-based kinematics is challenging with exoskeletons, as anatomical reference points, particularly on the pelvis, are occluded by exoskeleton structures. The aim of this study was therefore to develop and validate a method to reliably reconstruct the occluded pelvic markers. The movement data of six subjects, for whom pelvic markers could be placed while wearing an exoskeleton, were used to test the reconstructions and compare them to anatomical landmarks during lifting, holding and walking. Two separate approaches were used for the reconstruction. One used a reference coordinate system based on only exoskeleton markers (EXO), as has been suggested in the literature, while our proposed method adds a technical marker in the lumbar region (LUMB) to compensate for any shifting between exoskeleton and pelvis. Reconstruction with EXO yielded on average an absolute linear deviation of 54 mm ± 16 mm (mean ± 1SD) compared to anatomical markers. The additional marker in LUMB reduced mean deviations to 14 mm ± 7 mm (mean ± 1SD). Both methods were compared to reference values from the literature for expected variances due to marker placement and soft tissue artifacts. For LUMB 99% of reconstructions were within the defined threshold of 24 mm ±9 mm while for EXO 91% were outside.

Keywords:

• Exoskeleton
• motion capture
• kinematics
• anatomical markers

Authors’ contributions

Jasper Johns: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing—original draft. Alwina Bender: Conceptualization, Data curation, Formal analysis, Methodology, Writing—review & editing. Ulrich Glitsch: Conceptualization, Formal analysis, Methodology, Software, Supervision, Validation, Writing—review & editing. Lorenz Schmidt-Bleek: Data curation, Formal analysis, Investigation. Jörn Dymke: Conceptualization, Data curation, Investigation, Resources, Methodology. Christopher Brandl: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing—review & editing. Philipp Damm: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing—review & editing. Kai Heinrich: Conceptualization, Funding acquisition, Project administration, Supervision.

Disclosure statement

The authors report there are no competing interests to declare.

Additional information

Funding

This work was supported by the German Federal Ministry for Education and Research under the workHEALTH project grants 01EC1905B, 01EC1905C and 01EC1905D.