Mean-Square Exponential Consensus of Nonlinear Multi-Agent Systems via Distributed Random Impulsive Control

This paper aims to study mean-square exponential consensus of nonlinear leader-following multi-agent systems (MASs) in the presence of noise via distributed random impulsive control. The noise is driven by a second-order moment process with bounded mean power. A distributed random impulsive controller whose impulses occur at random moments is designed to ensure the desired consensus performance while enhancing security of the system. Sufficient conditions in terms of the network topology and impulsive parameter are derived, using the Lyapunov-based approach and properties of stochastic process. Finally, a numerical example is presented to demonstrate the effectiveness of our theoretical results.