Discrete-time function consensus algorithms for second-order moment statics

  • Yang Meng
  • , Tao Li
  • , Ji Feng Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper studies distributed function calculation by discrete-time consensus algorithms. We propose a class of discrete-time recursive algorithms to calculate the weighted second-order moment statics in a distributed way. Different from the continuous-time algorithm, the jumping of the states in discrete-time algorithms makes the convergence analysis much more difficult. In this paper, we first give some sufficient conditions on the control input for driving the whole system to function consensus. Then based on the conditions, we design the algorithm and show that if the communication topology is connected, then by adjusting the communication gain between adjacent agents properly, the proposed distributed algorithm can drive all agents' states to the square root of the weighted second-order moment statistics of the initial values. The numerical simulation shows the effectiveness of our algorithms.

Original languageEnglish
Title of host publicationProceedings of the 34th Chinese Control Conference, CCC 2015
EditorsQianchuan Zhao, Shirong Liu
PublisherIEEE Computer Society
Pages7240-7244
Number of pages5
ISBN (Electronic)9789881563897
DOIs
StatePublished - 11 Sep 2015
Externally publishedYes
Event34th Chinese Control Conference, CCC 2015 - Hangzhou, China
Duration: 28 Jul 201530 Jul 2015

Publication series

NameChinese Control Conference, CCC
Volume2015-September
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference34th Chinese Control Conference, CCC 2015
Country/TerritoryChina
CityHangzhou
Period28/07/1530/07/15

Keywords

  • Communication Gain
  • Distributed Algorithm
  • Function Consensus
  • Second-Order Moment Statistics

Fingerprint

Dive into the research topics of 'Discrete-time function consensus algorithms for second-order moment statics'. Together they form a unique fingerprint.

Cite this