A bionic approach for high-efficiency sensor data processing in building automation

Abstract

Abstract-Today's building automation shifts from simple applications like controlling illumination and temperature towards complex applications like the observation of buildings for safety and security reasons or increasing the occupants' comfort. Complex applications require more and more sensory information. Existing approaches of sensor data processing cannot cope with these demands. In this article, we present a new, bionically inspired approach for coping with such a huge amount of diverse sensory information. For this purpose, an information processing principle called neuro-symbolic information processing is introduced which bases on neuroscientific and neuro-psychological research findings about the perceptual system of the human brain.

Various Ways to Recall Cross-modality Perceptual Information in Cognitive Machine

“…In ideal environment setup which are well controlled and structured with finite and well-defined possible occurring situations, the observation and interpretation of the input conditions generally do not incur a serious problem to machine. Nevertheless, it is not the case in many real-world environments which are not well defined and controlled, comprising almost infinite number of possible occurring objects, events, and scenarios [2,3]. Hence, machine perception is required for more robust interpretation and subsequently, reasoning and action.…”