含有过时状态的二层邻居一致性算法

Consensus algorithm with outdated states of first-and second-order neighbors

为提高一致性收敛速度,将时滞作为积极因素考虑,提出一个含有过时状态的二层邻居一致性算法.与仅使用一层和二层邻居当前状态的标准二层邻居一致性算法相比,该算法不仅使用一层和二层邻居的当前状态,还使用其过时状态.在固定无向通信拓扑图下,对所提算法的稳定性和收敛性进行理论分析,结果表明:当过时状态选择恰当时,与标准二层邻居一致性算法相比,该算法具有更快的收敛速度.仿真结果验证了上述理论分析的有效性.

To improve the consensus convergence speed,we consider time delay as a positive factor and propose a distributed consensus algorithm for networked multi-agent systems in this paper.Com-pared with the standard consensus algorithm with the current states of first-and second-order neigh-bors,the proposed algorithm uses first-and second-order neighbors′both current states and outdated states stored in memory.Stability and convergence of the proposed algorithm were analyzed theoreti-cally under the undirected communication topology.It is shown that the proposed consensus algorithm converges faster than the standard consensus algorithm with the information of second-order neigh-bors,if the outdated states are chosen properly.Finally,simulation example was provided to demon-strate the effectiveness of the proposed algorithm.