Sensor-based assessment of ergonomic and kinematic implications of passive shoulder exoskeleton use

Abstract

Upper-limb musculoskeletal disorders are common in occupations involving repetitive or overhead tasks. Although passive shoulder exoskeletons are increasingly adopted in industrial settings to mitigate physical strain, their postural implications and changes in movement patterns remain underexplored. This study investigates the postural impact of two passive upper-limb exoskeletons (Ottobock Paexo Shoulder and Comau MATE-XT) during static and dynamic overhead tasks. Fifty-one participants were monitored using inertial motion tracking, and ergonomic risk was evaluated through Rapid Upper Limb Assessment (RULA). Exoskeleton use significantly altered upper-limb kinematics, with upper-arm elevation increasing by approximately 7–10° ( p  < 0.001), together with changes at the elbow and wrist, suggesting compensatory strategies. Mean RULA scores showed task- and side-dependent changes: in static tasks, scores changed from 4.24 to 3.96 on the dominant side and from 4.31 to 4.35 on the non-dominant side, whereas in dynamic tasks they significantly decreased from 5.28 to 5.09 and from 5.46 to 5.29, respectively ( p  < 0.05 after Holm–Bonferroni correction). Posture-only sensitivity analysis showed that this reduction was largely influenced by the Step-1 shoulder-support adjustment. Complementary linear mixed-effects models accounting for participant- and task-related variability did not retain significant fixed effects after correction. Gender-disaggregated and device-specific analyses revealed differences in postural adaptations. Overall, lower RULA scores coexisted with compensatory joint configurations diverging from biomechanically neutral postures. By integrating sensor-based kinematics with an ergonomic risk index, this work addresses a methodological gap and provides methodological and experimental evidence for occupational health, wearable robotics, and human-factors engineering.