Secure Output-Feedback Coordinated Maneuvering of Nonlinear Systems via Full-Order Neural Observer and Resilient Distributed Paths Update

This paper presents a secure output-feedback control method for coordinated maneuvering of nonlinear systems subject to immeasurable states, uncertain nonlinearities, timevarying disturbances, and denial of service (DoS) attacks. The attacks considered herein occur in the links rather than on the nodes. The proposed method includes a full-order state observer that observes states information, and a neural network that approximates the nonlinear uncertainties. An observer-based neural network adaptive control law is developed using a dynamic surface control approach to achieve path maneuvering. Besides a resilient distributed path update law is developed to enable distributed secure coordination under DoS attacks. The stability of the closed-loop system is established using an input-to-state cascade system analysis, and the secure coordination of path updates is analyzed. Finally, a simulation example is presented to demonstrate the effectiveness of the proposed method.