Abstract
Utilising a test house built to the author’s design in North London as its case study, this essay describes the need for reliable occupancy sensing in houses where automatic room-by-room environmental control takes place, before going on to discuss the complexity of transient room-by-room occupancy in this type of building. Doors are usually thought of as being kept closed in our homes. When however the doors are kept open, as is often the case, the house is identified as being generically different from a spatial, social and an environmental point of view. The essay makes the case for reliable local sensing on a room-by-room basis that is private and not dependent upon the Internet. Possible sensing devices are discussed, and the essay then goes on to analyse the combined use of PIR (passive infrared) and CO2 (carbon dioxide) sensors in the case study dwelling. The building is ventilated by a whole house heat recovery system (MHVR) and the results are provided for various conditions: a) doors and windows closed; b) doors open and windows closed; c) doors closed and windows open; and d) doors open and windows open. The paper concludes that in the first two of these conditions the combined sensors will robustly indicate occupancy in the house, and that local machine learning could support PIR-only sensing in the latter two conditions.