Abstract
In-home elderly care faces a crucial challenge regarding mobility among essential assistive devices, for which dual-arm care robots present a viable solution. However, ensuring human comfort in human-robot interactions necessitate quantifiable standards. Currently, the field lacks accurate biomechanical model solutions and objective comfort evaluation. In response to this need, this study proposes a method for solving human-robot statics models based on real-time pressure and position information. Employing the Optitrack motion capture system and Tekscan pressure sensors, we collect real-time positional and pressure data. This information is then incorporated into our human-robot statics model, facilitating the instantaneous calculation of forces and moments within the human body's sagittal plane. Building on this, comprehensive research literature review and meticulous questionnaire surveys are conducted to establish a comprehensive comfort evaluation function. To validate this function, experiments are performed to enable real-time assessment of comfort levels experienced during the process of transferring the human body. Additionally, the Noraxon surface electromyography (sEMG) sensors are utilized to capture real-time sEMG signals from the erector spinae, adductor muscles and quadratus lumborum, thereby providing objective validation for the comfort evaluation function. The experimental findings demonstrate that the proposed methodology for evaluating comfort achieves an accuracy rate of 85.1%.