Mimicking biological mechanisms for sensory information fusion
- Autori: DE PAOLA, A.; LA CASCIA, M.; LO RE, G.; Morana, M.; Ortolani, M.
- Anno di pubblicazione: 2013
- Tipologia: Articolo in rivista (Articolo in rivista)
- OA Link: http://hdl.handle.net/10447/72933
Current Artificial Intelligence systems are bound to become increasingly interconnected to their surrounding environment in the view of the newly rising Ambient Intelligence (AmI) perspective. In this paper, we present a comprehensive AmI framework for performing fusion of raw data, perceived by sensors of different nature, in order to extract higher-level information according to a model structured so as to resemble the perceptual signal processing occurring in the human nervous system. Following the guidelines of the greater BICA challenge, we selected the specific task of user presence detection in a locality of the system as a representative application clarifying the potentialities of cognitive models. Specifically, our contribution lies in the definition of a suitable model for knowledge representation and management; our goal is to make the artificial system able to understand the environment in which it acts, analogously to the way the human brain acts. In our system, the fusion of several information flows is performed by a Hidden Markov Model that allows to deal with heterogeneous data, potentially affected by a non-negligible degree of uncertainty, also taking into account the history of past perceptions. Sensory data are provided to the inference engine by a sensor network acting as a ‘‘peripheral nervous system’’ which performs a preliminary processing, thus mimicking a mechanism present in the human beings. Our cognitive approach to information fusion is not limited to the specific case study, but it can be easily generalizable to any context characterized by a striking heterogeneity in the sensory system. Promising results have been obtained during the assessment of the information fusion model which performed very satisfactorily in terms of specificity and sensitivity.