Predefined-Time Fuzzy Reinforcement Learning Control for Secure Surrounding Formation of NMSVs With DoS Attacks

This article studies the secure surrounding formation (SSF) problem of networked marine surface vehicles subject to denial of service (DoS) attacks. A hierarchical control framework is developed for designing the predefined-time fuzzy reinforcement learning controller, which consists of two layers. The distributed resilient estimator is proposed to accurately estimate the trajectory of the leader center in the predefined-time under DoS attacks over digraphs. The fuzzy reinforcement learning local controller is designed to achieve the SSF within the predefined-time. The sufficient conditions for system convergence and stability are derived based on the Lyapunov stability theory. Finally, simulation experiments are conducted to verify the effectiveness of the theoretical results.