Vehicular Mobility Prediction by Bayesian Networks

DAIMI Report Series

View Publication Info
 
 
Field Value
 
Title Vehicular Mobility Prediction by Bayesian Networks
 
Creator Schougaard, Kari
 
Description In mobile and ubiquitous computing the location of devicesis often important both for the behavior of the applicationsand for communication and other middleware functionality.Mobility prediction enables proactively dealingwith changes in location dependent functionality. In thisproject Bayesian networks’ ability to reason on the basis ofincomplete or inaccurate information is powering mobilityprediction based on a map of the street grid and the currentlocation and direction of the vehicle. We found that itis feasible to divide information of a map into smaller partsand generate a Bayesian network for each of these in orderto make mobility prediction based on localized information.This makes the information stored in the Bayesian networksmore manageable in size, which is important for resourceconstrained devices. Common sense knowledge of how vehiclemoves is feeded into the networks and enables themto make a good prediction even when no information of thevehicles mobility history is used. Experiments on real worlddata show that in an area statically divided into hexagonalcells of 200m in diameter, we get 80.54% accuracy whenusing localized Bayesian networks to predict which cell avehicle enters next.
 
Publisher Aarhus University
 
Contributor
 
Date 2007-01-01
 
Type info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
Peer-reviewed Article
 
Format application/pdf
 
Identifier http://ojs.statsbiblioteket.dk/index.php/daimipb/article/view/7220
10.7146/dpb.v36i582.7220
 
Source DAIMI Report Series; No 582 (2007): PB-582 Vehicular Mobility Prediction by Bayesian Networks
DAIMI Report Series; No 582 (2007): PB-582 Vehicular Mobility Prediction by Bayesian Networks
2245-9316
0105-8517
 
Language eng
 
Relation http://ojs.statsbiblioteket.dk/index.php/daimipb/article/view/7220/6160
 

Contact Us

The PKP Index is an initiative of the Public Knowledge Project.

For PKP Publishing Services please use the PKP|PS contact form.

For support with PKP software we encourage users to consult our wiki for documentation and search our support forums.

For any other correspondence feel free to contact us using the PKP contact form.

Find Us

Twitter

Copyright © 2015-2018 Simon Fraser University Library