Event-Based Reset Control for Consensus of Multiagent Systems Under Denial-of-Service Attacks: An Appointed-Time-Observer-Based Approach
Yuangui Bao, Jialing Zhou, Guanghui Wen, Dezhi Zheng, Zhongxiang Li
Abstract
This article studies the leader-following consensus of general linear multiagent systems (MASs) subject to denial-of-service (DoS) attacks via an event-based reset control protocol with only output information available. A local observer is constructed for each agent to exactly reconstruct the agent’s states at the appointed time, and a resilient event-triggered mechanism (RETM) is adopted to regulate the transmission of estimated states. Based on the estimated states, a novel reset control protocol composed of linear and reset parts is proposed, where the reset part improves the transient consensus performance. Compared with related results, the full state information of the agents is not required, continuous-time or periodic communication is avoided, and DoS attacks are tackled in this article. Furthermore, a hybrid impulsive model is developed to describe the dynamics of MASs under the reset control protocol, the ETM, and DoS attacks. By employing the hybrid system analysis approach, sufficient yet effective conditions are provided to guarantee secure consensus. Also, Zeno behavior is excluded under the reset strategy and ETM. Numerical simulations are conducted to verify the theoretical results.