Decentralized unknown input observer-based state and fault estimation for multi agent systems

This study addresses the challenge of decentralized state and fault estimation in Lipschitz nonlinear multi-agent systems. We designed unknown input observers, taking into account both system and output disturbances. In order to establish a local model for each agent, information is gathered from the agent itself and its neighboring agents. Consequently, an agent can robustly estimate not only its own state and faults but also those of its neighboring agents. Conditions for observer design are derived in terms of linear matrix inequalities. Numerical simulations are utilized to illustrate the effectiveness of the proposed approach.