Sensor fusion for a network of processes/systems with highly autonomous sensors

Sensor fusion systems have largely been implemented with centralized and hierarchical architectures using numerical and statistical inference methods. Some recent research has emphasized distributed and decentralized systems, using analytical/quantitative inference mechanism. It appears that little has been done to define generic paradigms and theories to apply qualitative reasoning as an inference mechanism in sensor fusion systems. This paper describes a distributed sensor-fusion paradigm and theory based on a previously developed theory to model sensors as highly autonomous units. Generic frameworks are defined to reason and make decisions at the qualitative level. Therefore, fusion is done at a high-qualitative level, thus making it possible for the implementation of intuitive yet effective methods to monitor, diagnose, and compensate processes and their sensors. This paradigm facilitates distribution of intelligence to the sensor level and sharing of information among sensors, controllers, and other devices in the system