De-RANSAC: Decentralized RANSAC for Sensor Networks
Eduardo Montijano, Sonia Martínez, and Carlos Sagüés
2010 American Control Conference, submitted
Abstract:
This paper studies the problem of distributed consensus in the presence of spurious sensor information. We propose a new method, De-RANSAC, which allows a multi-agent system to detect outliers---erroneous measurements or incorrect hypotheses---when the sensed information is gathered in a distributed way. The method is an extension of the RANSAC (RANdom SAmple Consensus) algorithm, which leads to a consensus result on the goodness of a set of measurements with certain probability. In order to execute the full process in a decentralized way, we propose a distributed voting policy valid for fixed and switching topologies. Simulations of real applications are provided showing the reliability of the proposed method.
File: main.pdf
Bib-tex entry:
@InProceedings{EM-SM-CS:10-acc,
author = {E. Montijano and S. Mart{\'\i}nez and C. Sag\"{u}\'{e}s},
booktitle = {Proceedings of the 2010 American Control Conference},
title = {De-{RANSAC}: {D}ecentralized {RANSAC} for {S}ensor {N}etworks},
year = {2009},
address = {Baltimore, USA},
note = {Submitted}
}