Multi-room occupancy estimation through adaptive gray-box models

We consider the problem of estimating the occupancy level in buildings using indirect information such as CO₂ concentrations and ventilation levels. We assume that one of the rooms is temporarily equipped with a device measuring the occupancy. Using the collected data, we identify a gray-box model whose parameters carry information about the structural characteristics of the room. Exploiting the knowledge of the same type of structural characteristics of the other rooms in the building, we adjust the gray-box model to capture the CO₂ dynamics of the other rooms. Then the occupancy estimators are designed using a regularized deconvolution approach which aims at estimating the occupancy pattern that best explains the observed CO₂ dynamics. We evaluate the proposed scheme through extensive simulation using a commercial software tool, IDA-ICE, for dynamic building simulation.