Real-time model-based fault detection and diagnosis for automated systems

Abstract

This paper describes a model-based fault detection and diagnosis approach for automated systems and ongoing developments in this area. The design is based on concepts of hierarchical control used in supervised autonomous robotic systems. The connection to automated systems design is first described, then the fault detection/diagnosis method is described. This method employs model-based fault detection, model-based fault diagnosis, and adaptive modelling. A new integrated fault detection/diagnosis approach is being developed using constraint propagation. The software used in these projects is an event-driven, component-based, real-time software environment developed for robot and automated system control applications. The component-based software approach simplifies model design and development, and the event-driven paradigm supports constraint propagation implicitly.