线性离散一致性系统均方偏差上界upperboundofmean-square.pdf

第 37 卷第 10 期 通 信 学 报 Vol.37 No.10 2016 年 10 月 Journal on Communications October 2016 doi:10.11959/j.issn.1000-436x.2016193 线性离散一致性系统均方偏差上界 窦全胜 1,2 ，厉玉蓉 1 ，姜平 1 (1. 山东工商学院计算机科学与技术学院，山东 烟台 264005 ； 2. 山东省车联网工程技术中心，山东 烟台 264005) 摘 要：一致性问题是多智能体协同控制的基础，有广泛的应用背景。之前的研究已经给出了线性一致性协议实 现均方一致的充分条件，在此基础上，对线性离散均方一致性问题进行了更加深入的讨论，以图论和随机分析为 基础，进一步给出并严格证明线性离散均方一致性问题噪声偏差的上界，同时以 4 种不同的网络拓扑为例，对所 提理论结果进行了说明。结论将为线性离散一致性系统的噪声估计和控制提供理论依据 关键词：群体智能；多智能体系统；一致性协议；噪声问题；均方一致性 中图分类号：TP301.6 文献标识码：A Upper bound of mean-square deviation of discrete linear consensus system 1, 2 1 1 DOU Quan-sheng , LI Yu-rong , JIANG Ping (1. School of Computer Science and Technology, Shandong Institute of Business and Technology, Yantai 264005, China; 2. Shandong Vehicle Networking Engineering Technology Research Center, Yantai 264005, China) Abstract: The consistency problem is the basis of cooperative control of multi-agents, and has wide application back- ground. The previous research has given the sufficient condition for the linear consistency protocol to realize mean-square consistency problem . Based on graph theory and random analysis, the noise-deviation upper bound of linear discrete mean-square consistency problem was strictly proved further, at the same time, four different network were used as examples to explain the theoretical results. The conclusion provides theoretical basis for the noise evaluation and con- trol