Combining Quantitative and Qualitative Models with Active Observations for better Diagnoses of Autonomous Mobile Robots

Quantitative and qualitative models and reasoning methods for diagnosis are able to cover a wide range of divers properties of a system. Both groups of methods have advantages and drawbacks in respect to fault diagnosis. In this paper we propose a framework which combines methods of both group to a combined diagnosis engine in order to improve the overall quality of diagnosis. Moreover, we present the different methods based on a running example of an autonomous mobile robots. Furthermore, we discuss the problems and research topics which arise from such a fusion of diverse methods. Finally, we explain how actively gathered observation are able to further improve the quality of diagnosis of complex systems.