LAUSR

This letter proposes a marking estimation approach for discrete event systems modeled by signal interpreted Petri nets (SIPN). The current marking of the SIPN is estimated based on the current sensor outputs and on the predicted markings inferred from the last estimate. This approach applies to any system that can be described by an SIPN and has no restriction on the indistinguishability of transitions or on the number of markings corresponding to the same sensor outputs. This letter also proposes a new fault detection scheme that uses the SIPN model of the fault-free system. Based on the results of the marking estimation, a decision logic is derived that can be used to detect sensor and actuator faults in the system.