An Integrated framework for combining global and local analyses in diagnosing hybrid systems

Sensor-rich systems typically employ extensive signal processing to detect and classify faults for fault isolation tasks. Sensor-poor systems, on the other hand, require models of the system structure and behavior and analytical redundancy techniques to make diagnostic inference. Because of the increasing availability of inexpensive, batch-fabricated micro-controllers and MEMS sensors, it is common for electro-mechanical systems such as office equipment and vehicles to deploy multitude of sensors and microprocessors for control and diagnosis tasks. Most of these systems tend to be embedded systems involving discrete control actions that determine the region of continuous operation of the device. We develop a diagnosis method that combines model-based diagnosis with signal processing techniques to address the challenges in diagnosing complex systems with hybrid discrete/continuous behaviors and to reduce the computational requirements by focusing the signal processing algorithms. We demonstrate the approach on problems in reprographic copier paper path diagnosis, and discuss the computational results.