In addition to developing FORK for the project Human-in-the-loop Sensing and Control for Commercial Building Energy Efficiency and Occupant Comfort (DOE#: DE-EE0007682), we launched a year-long, large-scale, human-in-the-loop Thermal Comfort Study (TCS). We wish to predict and evaluate the thermal comfort of 80 smart building occupants in a fully-sensed and controlled thermal chamber at Carnegie Mellon University.
This prediction, using a combination of different data-driven and thermal modelling methods, given environmental sensor data, and bio-metrics, will serve as a basis for generating subjective comfort models that will be used in autonomous HVAC system control as well as other aspects of user personalisation in the space. Occupants will be instrumented with wearable devices that sense and transmit bio-signal data, e.g., skin temperature, galvanic skin response, and metabolic heart rate. The user bio-signal data will be correlated with environmental sensor data, the user’s physical measures (e.g., manually and from depth cameras), and the user’s subjective responses on a mobile application (clothing level, subjective comfort level, activity, etc.).