Fault Diagnosis with Dynamic Observers

TL;DR

This paper explores the use of dynamic observers in fault diagnosis for discrete event systems, demonstrating polynomial-time diagnosability checking, a game-theoretic synthesis method, and optimization of observer costs.

Contribution

It introduces the concept of dynamic observers that can adaptively switch sensors, extending diagnosability analysis and synthesis methods beyond static observers.

Findings

Diagnosability for dynamic observers is polynomial-time checkable.

Most permissive dynamic observer can be synthesized in doubly exponential time.

Optimization of observer costs is formulated and analyzed.

Abstract

In this paper, we review some recent results about the use of dynamic observers for fault diagnosis of discrete event systems. Fault diagnosis consists in synthesizing a diagnoser that observes a given plant and identifies faults in the plant as soon as possible after their occurrence. Existing literature on this problem has considered the case of fixed static observers, where the set of observable events is fixed and does not change during execution of the system. In this paper, we consider dynamic observers: an observer can "switch" sensors on or off, thus dynamically changing the set of events it wishes to observe. It is known that checking diagnosability (i.e., whether a given observer is capable of identifying faults) can be solved in polynomial time for static observers, and we show that the same is true for dynamic ones. We also solve the problem of dynamic observers' synthesis and prove that a most permissive observer can be computed in doubly exponential time, using a game-theoretic approach. We further investigate optimization problems for dynamic observers and define a notion of cost of an observer.