Predictive Cloud Control for Networked Multiagent Systems With Quantized Signals Under DoS Attacks

Abstract

In this paper, a predictive cloud control scheme is investigated for a networked multiagent system (NMAS) with quantized signals under denial-of-service (DoS) attacks. An arbitrary region quantizer is applied to take effective values of control signals. A Stackelberg game is adopted to model the DoS attacks which result in attack-induced packet dropouts. Based on cloud computing and predictive control, the predictive cloud control scheme is designed to compensate for network-induced delays and the attack-induced packet dropouts actively. By a “zoom” strategy, sufficient conditions are given to achieve consensus for the NMAS. Effectiveness and computing superiority of the proposed scheme are demonstrated by simulation results.