AKRI

Intelligence :
Intelligent Homes

The original research project completed in 1990 was entitled 'Investigation into the Application of Artificial Intelligence to Small Building Management'. The original thesis is not available on line but a paper describing the work in less detail is available at ‘Machine Intelligence in a Human Environment’.

A more detailed description of the current Intelligent Home project can be found here.

The main interest of the original research was to investigate how much the system could learn about human occupancy patterns by using mainly simple movement sensors as input. The main function of the system was to try to establish how many people were in each room at any one time (occupant location monitor). This was done by reference to a short term memory of sensor data and heuristics to detect and correct errors in reasoning when they occurred. The function of long term memory was to try to predict how many people were likely to be in any room in the near future and to predict the general occupancy pattern at the start of each day. This information was used to influence the decisions made by the short term occupancy location monitor. All long term memories were given an environmental context using all data available.
An Autonomic Microprocessor Controller provided direct control of certain features of the installation using simple local control rules. However, the Higher Level Processor was able to change the detailed application of these rules by reference to the output of the occupancy location monitor system.

Content on this page requires a newer version of Adobe Flash Player.

Get Adobe Flash player

Content on this page requires a newer version of Adobe Flash Player.

Get Adobe Flash player