Adaptive neural event-triggered secure control for nonlinear MASs against actuator faults and FDI attacks
Yumeng Cao, Ben Niu, Huanqing Wang, Xudong Zhao, Abdullah Al-Barakati
Abstract
In this paper, an adaptive event-triggered consensus control scheme is designed for nonlinear multi-agent systems suffering from actuator faults and false data injection (FDI) attacks. To counteract the corruption of input states caused by sensor-level FDI attacks, a novel coordinate transformation is proposed, enabling accurate state reconstruction. By modifying local control laws and adaptive laws, the negative effects of actuator faults are counteracted. Unknown time-varying attack weights are addressed using Nussbaum-type functions. In addition, an adaptive event-triggering mechanism based on coordinate transformation is established to reduce the communication frequency of the network. The validity of the proposed control protocol is evidenced by two simulation examples.