Human-based models for smart devices in ambient intelligence

Over the last few years, there has been growing interest in ubiquitous computing and networking systems necessary to design environments in which users are surrounded by many items of equipment and sensors. To trigger appropriate services, it is essential 1) to discover the user's context and 2) to deliver adaptive services. Ambient Intelligence domain is a good arena to experiment these issues due the need to automatically detect context from wearable or environmental sensor systems and to transform such information for achieving personalized services. Here we describe an agent-based ambient intelligence architecture able to deliver services on the basis of physical and emotional user status captured from a set of biometric features. Abstract representation and management is achieved thanks to two markup languages, H2ML and FML, able to model behavioral as well as fuzzy control activities and to exploit distribution and concurrent computation in order to gain real-time performances. © 2005 IEEE.