Sensor Fault-Tolerant State Estimation by Networks of Distributed Observers

We propose a state estimation methodology using network of distributed observers. consider scenario in which the local measurement at each node may not guarantee system's observability. In contrast, ensemble all measurements does ensure that observability property holds. As result, we design observers such estimated vector computed by observer converges to and communicated estimates subset its neighborhood. The proposed scheme exploits sensor redundancy provide robustness against faults sensors. Under suitable conditions on redundant sensors, show it is possible mitigate effects class estimation. Simulation trials demonstrate effectiveness scheme.