A dynamical cognitive multi-agent system for enhancing ambient intelligence scenarios

Ambient Intelligence (AmI) is born as a computer paradigm that deals with a new world where computing devices are spread everywhere in order to make wider the interaction between human beings and information technology and put together a dynamic computational-ecosystem capable of satisfying the users requirements. However, the AmI systems are more than a simple integration among computer technologies, indeed, their design can strongly depend upon psychology and social sciences aspects able to describe and analyze the human being status during the system's decision making. Consequently, from a computational point of view, an AmI system can be considered as a distributed cognitive framework composed by a collection of intelligent entities capable of modifying their behaviours by taking into account the user's cognitive status in a given time. This paper introduces a novel methodology of AmI systems' design that exploits multi-agent paradigm and a novel extension of Fuzzy Cognitive Maps theory benefiting on the theory of Timed Automata in order to create a collection of dynamical intelligent agents that use cognitive computing to define actions' patterns able to maximize environmental parameters as, for instance, user's comfort or energy saving. ©2009 IEEE.