DoS Attacked Topology Linear Data Dynamic Modeling and Predictive Data Compensation based Data-Driven Consensus Control for Multi-agent Systems

This work investigates the issue of Denial-of-Service (DoS) attacks in nonaffine nonlinear multiagent systems (MASs). Firstly, an attack compensation scheme encompassing both input and output is designed. Subsequently, a DoS attacked topology linear data dynamic model (DoS-TLDM) is established leveraging the dynamic linearization method among agents and the network topology. Finally, a data-driven consensus control based on predicted data compensation (PDC-DDC) is presented by leveraging the DoS-TLDM and the attack compensation scheme. The PDC-DDC belongs to a data-driven method. The simulation results validate the operational efficacy of the proposed methodology.