Department of Computer Science and Institute of Cognitive Science
University of Colorado
|Topic:||The Adaptive House|
|Date:||Thursday, January 28, 1999|
|Place:||Gould-Simpson, Room 701|
Although the prospect of computerized homes has a long history, home automation has never become terribly popular because the benefits are seldom seen to outweigh the costs. One significant cost of an automated home is that someone has to program it to behave appropriately. Typical inhabitants do not want to program simple devices such as VCRs, let alone a much broader range of electronic devices, appliances, and comfort systems that have even greater functionality. We describe an alternative approach in which the goal is for the home to essentially _program itself_ by observing the lifestyle and desires of the inhabitants, and learning to anticipate and accommodate their needs. We have constructed a prototype system in an actual residence using neural network and reinforcement learning techniques. The residence is equipped with sensors to provide information about environmental conditions (e.g., temperature, ambient lighting level, sound, and motion) and actuators to control basic residential comfort systems---air heating, lighting, ventilation, and water heating. By predicting lifestyle patterns of the residents, the system can infer rules of operation that anticipate inhabitant needs while conserving energy.