Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/45879
Title: Dempster-shafer theory based ship-ship collision probability modelling
Authors: Talavera Ortiz, Alejandro 
Aguasca Colomo, Ricardo 
Galván González, Blas J.
UNESCO Clasification: 3307 Tecnología electrónica
Keywords: Maritime traffic safety
Dempster-Shafer Theory
collision probability
Data AIS
Issue Date: 2013
Journal: Lecture Notes in Computer Science 
Conference: 14th International Conference on Computer Aided Systems Theory (EUROCAST) 
14th International Conference on Computer Aided Systems Theory, EUROCAST 2013 
Abstract: The methodology proposed in this paper considers the uncertainty present in modeling the probability of collision between ships on a route. The proposal allows representing and quantifying uncertainty, and ensures rigorous propagation of this uncertainty from the input variables to the output variable. This proposal complements the analysis of risk and helps the decision maker to know the degree of confidence associated with the results of the analysis. Pedersen’s model has been selected to estimate the probability of collision, using the information provided by the AIS, and Dempster-Shafer Theory has been selected for the treatment of uncertainty. This methodology has been applied to maritime traffic in the Canary Islands and has been validated using the Kullback-Leibler divergence. The results are consistent with those obtained with the software IWRAP recommended by IALA.
URI: http://hdl.handle.net/10553/45879
ISBN: 9783642538612
ISSN: 0302-9743
DOI: 10.1007/978-3-642-53862-9-9
Source: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)[ISSN 0302-9743],v. 8112 LNCS, p. 63-70
Appears in Collections:Actas de congresos
Show full item record

SCOPUSTM   
Citations

3
checked on Feb 21, 2021

WEB OF SCIENCETM
Citations

1
checked on Feb 21, 2021

Page view(s)

38
checked on Feb 21, 2021

Google ScholarTM

Check

Altmetric


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.