Adaptive Neural Self-Triggered Bipartite Secure Control for Nonlinear MASs Subject to DoS Attacks

Fabin Cheng,Hongjing Liang,Ben Niu,Ning Zhao,Xudong Zhao
DOI: https://doi.org/10.1016/j.ins.2023.02.058
IF: 8.1
2023-03-01
Information Sciences
Abstract:This paper studies the bipartite secure control design problem for nonlinear multi-agent systems (MASs) subject to denial-of-service (DoS) attacks over a signed digraph. By proposing an anti-attack control method, the nonlinear MASs can achieve the bipartite secure control goal in an insecure network and physical environment. In addition, to save communication resources, a novel distributed adaptive self-triggered control (ASTC) mechanism is proposed. Different from the traditional self-triggered control, the trigger interval can be dynamically adjusted according to the situation that the followers converge to the convex hull formed by the leaders, which makes the proposed control protocol based on ASTC able to balance the system performance and communication resources under DoS attacks through scheduling the inherent system resources. Furthermore, in order to overcome the problem of complexity explosion, a command filter is introduced into the design process to simplify the design process. The validity of our control scheme is demonstrated through a simulation example.
computer science, information systems
What problem does this paper attempt to address?