Buildings as intelligent autonomous systems: a model for integrating personal and building agents

摘要

We describe a new application domain for intelligent autonomous systems – intelligent buildings. We show how a building can be regarded as a machine and how behaviour-based principles first proposed by Brooks for mobile-robot control can be applied to enable autonomous intelligent-building agents to adapt their control to suit the occupants. We present a novel approach to the implementation of Intelligent Buildings based on a multi embedded-agent architecture comprising a low-level behaviour based reactive layer together with a high- level deliberative layer based on evidential learning (a case-like learning mechanism). We also present a hierarchical agent architecture in which mobile agents (residing on body wearable devices) and fixed agents (residing in buildings) can be integrated, opening new commercial and personal possibilities. We discuss how this architecture is being implemented, using a combination of IP and Lonworks networking technology together with a Java programming environment. We consider future directions of this work, in particular how it might play a key role in intelligent interactive environments enabled by emerging technologies such as mobile phones and embedded-internet devices.