Cognitive Smart Environment: an approach based on Concept Hierarchies and Sensor Data Fusion

Sensor data fusion gathers technological solutions which are devoted to collecting, classifying and combining data from multiple (and possibly diverse) sensors deployed in a Smart Environment in order to augment knowledge about the system so realizing cognitive behavior. The goal is that of making more effective the analysis and management of acquired data as well as that of promoting the realization of cognitive systems which can sense the environment, reason about what is happening around it and properly act for reaching some purposes. Anyway, the development of such cognitive smart environments asks for suitable approaches able to jointly face with issues like heterogeneity, sensor/actuator management, system reactivity, behavioral modeling, and context awareness. This paper proposes a multi-tier approach, based on concept hierarchies and sensor data fusion, dealing with the above aspects in a modular way. The approach favors separation of concerns and abstractions. It relies on the use of the agent metaphor and statecharts. The use of the iSapiens IoT-based platform is then suggested for development purposes. As a significant case study, the approach was exploited for the development of a Smart Museum by making cognitive some artworks (statues) of the MAB open-air museum located in the city of Cosenza, Italy.