Observer-based non-fragile H∞-consensus control for multi-agent systems under deception attacks

This paper addresses the non-fragile $ H_{\infty } $ H∞-consensus control problem for multi-agent systems subject to deception attacks and missing measurements based on the designed observers. First of all, a sequence of Bernoulli distributed stochastic variables is applied to characterise the random nature of the missing measurements, where the changeable missing probability is constrained by a norm-bounded condition to reflect the reality more closely. Next, constraint conditions are given for the malicious signal considered in deception attacks with the purpose of limiting the difference between the attack signal and the true signal. The aim of this research is to design an observer-based non-fragile $ H_{\infty } $ H∞-consensus controller for each agent such that the multi-agent closed-loop system can achieve the stochastic stability and meet the prescribed disturbance attenuation level when encountering the deception attacks and missing measurements. Then, based on Lyapunov stability theory, the sufficient condition is established to ensure the existence of the desired observer and controller. Moreover, the gain parameters are calculated by solving linear matrix inequalities. Finally, an explanative example is employed to demonstrate the effectiveness of the proposed distributed control algorithm.