A Questionnaire-Free Approach for Thermal Comfort Measurement Using Unsupervised Machine Learning on Physiological Signals

Traditional methods for personalized thermal comfort measurement rely on subjective questionnaires, which are often impractical to collect in real-world contexts due to their intrusiveness, discontinuity, and dependency on the user. To overcome these limitations, this study proposes an innovative methodology for the assessment of thermal comfort that measures solely physiological data, thereby eliminating the need for feedback collection. A nonintrusive sensing system comprising a wearable smartwatch, an infrared (IR) temperature sensor, and a social robot has been employed to measure occupants' skin temperature and heart rate (HR) variability under varying indoor thermal conditions. K -means clustering was applied to unlabeled physiological features. A silhouette score of 0.52 has been obtained with physiological data collected during the winter season measurement campaign, revealing that it is possible to discern comfort and discomfort sensation based solely on physiological signals. The results demonstrate the feasibility and robustness of a physiology-based, label-free approach for thermal comfort measurement, which provides a scalable alternative to survey-based methods.