Detecting occupancy and social interaction via energy and environmental monitoring

The demand for human oriented services in indoor environment has received steady interest and it is represent a big challenge for increasing the human well-being. In this work, we present a system able to perform room occupancy detection and social interactions identification, using data coming from both energy consumption information and environmental data. We also study the application of supervised and unsupervised learning techniques to the reference scenario, in order to: i) infer context information related to socialisation aspects, by recognising in real-time social interactions; ii) identify when a room is really occupied by workers or not, for emergencies management. The system has been tested in a real workplace scenario, inside three rooms of the CNR research area in Pisa occupied by different numbers of workers, representing the main core technology for future active and assisted living services.