Robust Unknown Input Observer-Based Fault Estimation of Leader–Follower Linear Multi-agent Systems

Abstract

In this paper, the problem of the robust unknown input observer for a class of linear multi-agent systems and its application to fault estimation are considered. First, an undirected graph is used to represent the communication topology of a leader---follower linear multi-agent system. Then, using relative output estimation errors among agents, an unknown input observer is proposed to achieve fault estimation for the global augmented system in which the actuator or sensor fault vector is taken as an auxiliary state vector. A multi-constrained design algorithm based on linear matrix inequality technique is also designed to obtain gain matrices of unknown input observer. Simulation results show the effectiveness and advantages of the proposed robust unknown input observer method for fault estimation of multi-agent systems.