PI: Bonny Banerjee
Hard and soft sensors that monitor a human or a facility around the clock are everywhere. Due to the sheer amount of data these sensors can generate, the resources required to safely store and analyze them are enormous. Since noteworthy events happen only occasionally in any environment, it is imperative for smart sensors to learn the norms in such data and make appropriate decisions at the occurrence of an abnormal event. Researchers at the Computational Intelligence Laboratory, affiliated with IIS and EECE, are developing a unified computational architecture for learning the norms from streaming data in multiple modalities in an unsupervised and online manner. The architecture is being implemented as an artificial agent that will raise an alarm in the presence of an abnormal object, action, or event in an environment. Multiple applications using the agent in security and surveillance, defense, and healthcare are being investigated.