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