Title
Distributed Data Association in Robotic Networks With Cameras and Limited Communications
Abstract
We address the data association problem of features that are observed by a robotic network. Every robot in the network has limited communication capabilities and can only exchange local matches with its neighbors. We propose a distributed algorithm that takes these local matches and, by their propagation in the network, computes global correspondences. When the algorithm finishes, each robot knows the correspondences between its features and the features of all the other robots, even if they cannot directly communicate. The presence of spurious local correspondences may produce inconsistent global correspondences, which are association paths between features observed by the same robot. The contributions of this study are the propagation of the local matches and the detection and resolution of these inconsistencies. We formally prove that after executing the algorithm, all the robots finish with a data association that is free of inconsistencies. We provide a fully decentralized solution to the problem that is valid for any fixed communication topology and with bounded communications between the robots. Simulations and experimental results with real images show the performance of the method considering different features, matching functions, and robotic applications.
Year
DOI
Venue
2013
10.1109/TRO.2013.2273843
IEEE Transactions on Robotics
Keywords
Field
DocType
image fusion,multi-robot systems,robot vision,bounded communications,cameras,communication topology,distributed algorithm,distributed data association,global correspondences,limited communication capabilities,network propagation,robotic networks,spurious local correspondences,Multi-agent systems,multi-robot systems,robot vision systems
Computer vision,Robot vision,Image fusion,Computer science,Distributed algorithm,Data association,Artificial intelligence,Real image,Robot,Spurious relationship,Bounded function
Journal
Volume
Issue
ISSN
29
6
1552-3098
Citations 
PageRank 
References 
10
0.54
20
Authors
3
Name
Order
Citations
PageRank
Eduardo Montijano121422.27
Rosario Aragues21179.18
Carlos Sagüés344339.22