Making Sense of Nonsense

The emergence of networked embedded systems and sensor/actuator networks has made possible the collection of large amount of real-time data about a monitored environment. Depending on the application, such data may have different characteristics: multidimensional, multi-scale, spatially distributed, time series. Moreover, the data values may be influenced by controlled variables, as well as by external environmental factors. However, in many cases the collected data may be incomplete, or it may not make sense for various reasons, thus compromising the sensor-environment interaction and possibly affecting the ability to manage and control key variables of the environment. Such problems are generally the result of some fault in the sensor/actuator system itself or an abnormality in the monitored environment, which may be either permanent or temporary, developing abruptly or incipiently. These problems become more pronounced as sensing/actuation systems get older. The main objective of this project is to develop intelligent methods for analyzing and interpreting the data such that faults are detected, isolated and identified as soon as possible, and accommodated for in future decisions or actuator actions. The problem becomes more challenging when these sensing/actuation systems are used in a wide range of environments which are not known a priori and, as a result, it is unrealistic to assume the existence of an accurate model for the behavior of various components in the monitored environment. Therefore, this project will focus on cognitive system approaches that can learn characteristics or system dynamics of the monitored environment and can adapt their behavior and predict missing or inconsistent data to achieve fault tolerant monitoring and control.