Abstract

For a networked swarm system with a directed path graph, this paper investigates a group evasive attack strategy against actuators from the attacker’s perspective. The core objective of the proposed strategy is to force the global network state to converge to a pre-specified synchronization trajectory. First, the evasive attack signal with an ecosystem-based generation mechanism is modeled, from which the pre-specified synchronization trajectory can be derived. Subsequently, an evasive attack protocol is designed by superimposing the developed evasive attack signals onto the nominal synchronization protocol of the networked swarm system. By exploring the projection of evasive attack signals onto the synchronization subspace, an explicit expression of the pre-specified synchronization trajectory is determined, which depicts the global network state of the system under group evasive attacks. Then, to mitigate the impacts of evasive attacks on the synchronization performance of the networked swarm system, a resilient framework integrated with robust H∞ regulation mechanisms is constructed to derive the design criteria for the group evasive attack strategy. Finally, a numerical simulation example is conducted to demonstrate the validity of theoretical results.

IPC Classification

Keywords