Consensus of nonlinear multiagent systems with uncertain disturbance under stochastic switching topologies

Abstract

In this paper, we investigate the mean square consensus problem of leader‐following nonlinear multiagent systems which suffer from white noise disturbance under randomly switching interaction topologies. The control protocol of each agent is designed based on its local information from its neighbors and the leader. Using matrix inequality theory and stochastic stability theory, a sufficient condition is presented to guarantee the mean square consensus. Moreover, two special cases: balanced topologies switching randomly and random connection failures are considered. Finally, two corresponding numerical simulations are provided to demonstrate the theoretical results.

Distributed consensus tracking of nonlinear MASs under stochastic switching topologies via aperiodically intermittent control

“…However, due to the decentralized and open structure, the communication network among agents may be affected by emergency situations in reality, such as external interference, random link failures, man-made attacks, and so forth. In this case, the interaction topologies can be described as the switching topologies, [31][32][33][34][35][36] and the robustness and reliability need to be investigated in depth. The asymptotic consensus problem was addressed by stochastic stability, in which all communication channels were independent of each other and switched randomly according to probability.…”

“…Define  =  + . Lemma 1 (35). If  is a strongly connected communication topology, then the corresponding matrix  and  T are both nonsingular matrices and all the eigenvalues are positive.…”

“…An adaptive consensus tracking strategy was proposed for a class of second‐order nonlinear MASs with time‐varying communication topology 34 . Under switching topologies determined by the Markov process, the mean‐square consensus problem was investigated for a class of nonlinear MASs with additive stochastic noise modeled by the white noise 35 . It should be noted that the previous control protocols are mainly designed for continuous control, which is not cost‐effective.…”

Necessary and sufficient conditions for mean‐square bounded consensus of multiagent systems with additive noise

“…In order to control the research cost and solve the problem of large number and complex structure of agents, an unbiased gradient technique is proposed in [15]. In addition, time delay and switching topologies will also affect the consensus of MASs [16][17][18].…”

Finite‐time and fixed‐time adaptive consensus of multi‐agent systems with general linear dynamics

“…In Xie and Mu [11], the leader-following consensus problem of nonlinear singular MASs with intermittent communication on fixed undirected topology was investigated. The mean square consensus problem of leader-following nonlinear MASs was investigated in Yang et al [12]. In recent years, more and more literatures have investigated consensus problem of general linear MASs.…”