A simple method has been suggested for the estimation of the occupied and unoccupied distributions for different sensors installed in a residential home.
A critical feature of the method is that it does not require extensive recording of ground truth.
Practical occupancy inference through combining Dempster-Shafer’s theory of evidence with Hidden Markov Models has been demonstrated on some preliminary data and appears to be a very reasonable approach.
A methodology has been developed that uses this practical occupancy inference for assessing the possibility of demand response for a particular household at different times of day.
The benefits of occupancy to different demand response initiatives have been qualitatively assessed.