Paper Info
Title
Consensus in Networks of Multiagents with Stochastically Switching Topologies and Time-Varying Delays.
Abstract
In this paper, we investigated a very general model of discrete-time consensus algorithm in networks of multiagents with stochastically switching topologies and bounded time-varying delays. Here not only the coupling structure is more general than existing works but the switching topologies are also modelled as general adapted random sequences, which do not necessarily possess any stationary or ergodic properties. To the best of our knowledge, this is the first work that simultaneously considers both stochastic switching topologies and time delays in the analysis of consensus algorithms. We first developed some new results to deal with the products of the stochastic matrices which arise in the analysis and do not have all positive diagonal entries. These results greatly improve some existing ones on the same kind of matrices, and they also extend a previous result on matrices with all positive diagonal entries. Then, we also developed some new results on random graph sequences to handle the subtleties induced by the stochastically switching structures. Based on these results, we proved essentially the same sufficient conditions for almost sure consensus as in our pervious works which are periodic jointly connectivity of the interaction graphs in conditional expectation. At last, we showed that beyond merely providing a more general framework, the significance of this work also lies in that it makes it possible to analyze in a random setting a class of consensus algorithms in networks with continuous-time dynamics but asynchronous discrete-time communications/updates which are important not only to sampled-data consensus algorithms but also to event/self-triggered ones.
Year
DOI
Venue
2018
10.1137/16M1095354
SIAM JOURNAL ON CONTROL AND OPTIMIZATION
Keywords
Field
DocType
consensus,multiagent systems,switching topology,adapted process,time delays
Diagonal,Mathematical optimization,Random graph,Matrix (mathematics),Adapted process,Ergodic theory,Network topology,Multi-agent system,Mathematics,Bounded function
Journal
Volume
Issue
ISSN
56
3
0363-0129
Citations 
PageRank 
References 
2
0.36
0
Authors
4
Name
Order
Citations
PageRank
Bo Liu119811.13
Wenlian Lu2133193.47
Licheng Jiao35698475.84
Tianping Chen43095250.77