Uncertainty Estimation of Human-Exoskeleton Interaction Using a Robotic Dummy

Exoskeletons and exosuits have witnessed unprecedented growth in recent years, especially in the medical and industrial sectors. The study of human-exoskeleton interaction relies on the participation of potential exoskeleton users, which gives rise to safety concerns and substantial testing resources. Specifically, the study of physical human-exoskeleton interaction (pHEI) is based on several metrics that are typically challenging to assess in a systematic way. To address this, we present an instrumented setup composed of a mechatronic leg able to sense interaction forces, together with a vision system for measuring pHEI kinematic metrics. In this study, we started from a previously designed protocol to extract key metrics utilized in assessing pHEI, such as joint misalignments, relative motions and interaction forces. Subsequently, we carried out an uncertainty analysis on these chosen metrics of interest. To achieve this, a series of experiments were conducted to assess the influence of the measurement uncertainty on each metric through multiple test repetitions. Despite generally acceptable uncertainty values, angular misalignment metrics warrant further investigation since their associated uncertainty is close to the value of its standard deviation in the conditions tested. These findings emphasize the importance of conducting such an analysis, offering insights for the future measurement protocols verification and improvements.