The application of principles coming from RCM (Reliability Centered Maintenance) in the aim of characterizing the behavior of systems where a human operator is included in the control loop. An approach for the selection of candidate variables able to describe the state of the system based on the study of interactions between the elements building the system and its environment is described. MCA is the proposed method to explore the set of candidate variables in order to identify' the most discriminant variables and also to put forward behavior classes.A first validation step of this approach is presented through its use to study evolutions of driver's behavior in a simulated motorway monotonous driving context. The method has shown its capacity to put forward time sensitive variables and their joint evolutions allowing to distinguish the different states of the system.
Using failure detection and diagnosis methods to detect dangerous evolutions of the driver behaviour
Anwendung von Fehlererkennungs- und Fehlerdiagnosemethoden bei der Überwachung des menschlichen Verhaltens während der Fahrzeuglenkung
2000
4 Seiten, 5 Bilder, 4 Quellen
Aufsatz (Konferenz)
Englisch
A Fuzzy Diagnosis Process to Detect Evolutions of a Car Driver's Behavior
British Library Conference Proceedings | 1994
|Driver dangerous driving detection and warning method
Europäisches Patentamt | 2023
|