Robust Security Control of a Class of Second-Order Nonlinear Systems Against DoS Attacks
针对受扰二阶非线性系统在拒绝服务攻击下的输出反馈安全跟踪控制问题，设计了基于径向基函数神经网络的有限时间状态观测器和自适应滤波器，并构建安全控制器，通过李雅普诺夫分析证明系统输出跟踪的一致最终有界性。
This article is concerned with the output feedback security tracking control of a class of disturbed second-order nonlinear systems against denial-of-service (DoS) attacks. Novel radial basis function neural network (RBFNN)-based finite-time state observers are developed to estimate the system’s unavailable states. Adaptive filters are proposed to suppress the influences of disturbances and RBFNN approximation errors. Then, an RBFNN-based security controller is designed to alleviate the effects of nonlinear dynamics and DoS attacks based on the signals of observers and filters. It is established that the uniformly ultimately bounded output tracking results of the system can be obtained by utilizing an RBFNN-based finite-time observation and filtering compensation control designs through Lyapunov stability analysis. Comparative simulations are employed to display the feasibility and superiority of the designed RBFNN-based observation and filtering compensation control schemes of a nonlinear autonomous marine system (AMS).