Remote Diagnosis of Timed I / O-Automata

This paper considers the consistency-based remote diagnosis of discrete-event systems, where the diagnostic task is decomposed into an onboard and an off-board task. The system to be diagnosed is modelled by a new type of timed automata possessing discrete inputs and outputs. Faults are modelled as parameters of the edges and invariants of the automata. A consistency-based diagnostic method is derived for this model class by extending a method developed for timed automata without inputs and outputs. This method can be used for fault detection on the on-board component and for fault identification on the off-board component of the remote diagnostic system. The results are illustrated by an example.