Affiliations: Automation Systems Group, Vienna University of Technology, Vienna, Austria | [x] University of Central Greece | [y] Democritus University of Thrace, Greece | [z] Frederick University, Cyprus
Abstract: The field of home and building automation is an extensive technological application area that inherently includes a high diversity and a strong interaction with users. Concepts like ubiquity and cooperativeness are of undeniable importance and demand a coherent top-down focus and holistic perspectives. The present paper describes the application of a global system concept for smart homes in the field of air quality and thermal comfort management. The proposed approach faces the high heterogeneity of smart homes by means of an extensive knowledge base and a cooperative framework of intelligent agents. On this structure, the underlying methodology identifies users' habits and behaviors as the main common elements that join home applications, services and technologies in a sound overall performance. The capabilities of context awareness reached by a holistic design allow the definition of an intermediate pre-control phase intended to make smart, optimized system-global decisions before the application of specific end controllers and devices.
Keywords: Smart homes, air quality, thermal comfort, Ambient Intelligence, user profiles, habit abstraction, multi-agent systems, decision making, machine learning
