Simultaneous fault detection and consensus control of linear multi-agent systems via reduced-order observer

Abstract

This paper investigates simultaneous fault detection and model-reference consensus control of multi-agent systems with reduced-order observers under the directed topology. A reduced-order observer based on the relative output information is designed, based on which the fault detection task is accomplished. By reconfiguring the observer matrices and designing the adaptive law to estimate the sensor fault, the control protocol is designed with only the relative output information and the model-reference consensus is achieved subject to the sensor fault. Simulation results illustrate the effectiveness of the our proposed design methodology.