H∞ performance tracking and group consensus of delayed multiagent systems under attack

Abstract

This paper deals with the problem of performance tracking and group consensus control of a multiagent system (MAS) under deception attack in the frequency domain. Deception attack, a type of network attack, is modeled and analyzed in a MAS where each agent is a continuous-time system. In contrast to prior works on group consensus, modeling of the attack and its impacts on the MAS are analyzed in the frequency domain. To overcome these effects, an H ∞ robust controller is proposed using the internal stability method. The proposed controller is optimized analytically for H ∞ performance tracking of the MAS and mitigation of deception attack simultaneously. Unlike attacks on communication links, this scheme considers attacks on nodes. The impacts of attacks on the MAS are eliminated without the removal of the compromised nodes. A numerical example demonstrates the effectiveness of the proposed scheme.