An adaptive fuzzy learning mechanism for intelligent agents in ubiquitous computing environments

We present a novel system for learning and adapting fumy controllers for intelligent agents embedded in ubiquitous computing environments supporting the user's activities. We have performed unique experiments where the intelligent agent learn and adapted online to the user's behaviour, while staying for a period of five days in the Essex intelligent Dormitory (iDorm) (iDorm) which is a real ubiquitous computing environment test bed. Offline and online experimental results are presented comparing the performance of our technique with other soft computing approaches. The results show that our system outperformed the other systems while operating online in a life long learning mode.