Active Disturbance Rejection Control to Consensus of Second-Order Stochastic Multiagent Systems

In this article, we attempt for the first time to apply the active disturbance rejection control approach to the disturbance rejection and consensus for a class of second-order stochastic multiagent systems, whose network topology is undirected and connected. Extended state observers are designed for online estimation of the unmeasured states and random total disturbance, which is the effect of external nonrandom disturbance, colored noise, and bounded noise corrupting each agent. Active antidisturbance consensus protocols based on the estimates of unmeasured states and random total disturbance are then designed to guarantee the consensus in both mean square and almost sure practical sense, while random total disturbance of each agent is compensated in real time in the closed loop. Some numerical simulations are performed for a practical system to verify the rationality of the consensus protocols.