Distributed fault estimation for multi-agent systems using relative measurements — Distributed adaptive observer approach

Abstract

In this paper, a distributed adaptive fault estimation strategy for multi-agent systems based on the relative output measurements is proposed. First, a novel fully distributed observer is designed in which relative state is estimated only having access related output and neighbors input signal. Then using the distributed observer, an adaptive fault estimator is proposed based on the relative output estimation error. It is provided that the relative state observer error is asymptotically stable. Moreover, similar to the existing works a transformed form of the fault estimation error is proven to be asymptotically stable which is drawn from the observable subspace of network dynamics. This transformation is related to the network topology. The results on the stability of the proposed theorems are in LMI form which is easily solvable in comparison to existing works. Finally, a numerical example is given to verify the effectiveness and superiority of the proposed method using only relative output measurements.