HIERARCHICAL MODEL-BASED DIAGNOSIS FOR HIGH AUTONOMY SYSTEMS

Deep reasoning diagnostic procedures are model-based, inferring single or multiple faults from the knowledge of faulty behavior of component models and their causal structure. The overall goal of this paper is to develop a hierarchical diagnostic system that exploit knowledge of structure and behavior. To do this, we use a hierarchical architecture including local and global diagnosers. Such a diagnostic system for h igh autonomy systems has been implemented and tested on several exampl es in the domain of robot-managed fluid-handling laboratory.