dc.contributor.author |
Mason, Andrew |
en |
dc.contributor.author |
Henderson, SG |
en |
dc.contributor.editor |
Sundaramoorthi, D |
en |
dc.contributor.editor |
Lavieri, M |
en |
dc.contributor.editor |
Zhao, M |
en |
dc.coverage.spatial |
Texas, USA |
en |
dc.date.accessioned |
2012-04-02T23:54:22Z |
en |
dc.date.issued |
2010-11-06 |
en |
dc.identifier.citation |
5th INFORMS Workshop on Data Mining and Health Informatics (DM-HI 2010), Texas, USA, 06 Nov 2010 - 06 Nov 2010. Editors: Sundaramoorthi D, Lavieri M, Zhao M. Proceedings of the 5th INFORMS Workshop on Data Mining and Health Informatics (DM-HI 2010). INFORMS. 10 pages. 06 Nov 2010 |
en |
dc.identifier.uri |
http://hdl.handle.net/2292/16513 |
en |
dc.description.abstract |
The Optima Corporation develops simulation systems for ambulance operators. These systems require careful calibration of road speeds under both standard and “lights and sirens” driving conditions. These speeds can often be estimated using GPS position data that is transmitted by the ambulances. However, this data is typically sparse in that it has large distances and/or time intervals between successive data points, and so determining likely vehicle routes is difficult. To address this, we have developed a dynamic programming optimisation algorithm that determines the most likely route taken by a vehicle using a sequence of recorded GPS locations and times. Unlike other more ad-hoc approaches, we formally define and then solve a complex maximum a-posterior probability problem that takes into account data such as the vehicle's location, heading, speed, and its “lights and sirens” status. This algorithm explores many different possible candidate vehicle locations for each recorded GPS data point. Preference is given to shorter, more direct final routes that reflect likely driver behaviour. The algorithm is permitted to label GPS data points as erroneous in which case they have minimal influence on the final route. Results are presented for several case studies using historic ambulance data. |
en |
dc.format.medium |
CD |
en |
dc.publisher |
INFORMS |
en |
dc.relation.ispartof |
5th INFORMS Workshop on Data Mining and Health Informatics (DM-HI 2010) |
en |
dc.relation.ispartofseries |
Proceedings of the 5th INFORMS Workshop on Data Mining and Health Informatics (DM-HI 2010) |
en |
dc.rights |
Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated. Previously published items are made available in accordance with the copyright policy of the publisher. |
en |
dc.rights.uri |
https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm |
en |
dc.subject |
Ambulances, Map Matching, Dynamic Programming |
en |
dc.title |
An Optimisation Approach for Map Matching using Sparse Ambulance GPS Data |
en |
dc.type |
Conference Item |
en |
dc.description.version |
Obtain full text from author |
en |
dc.rights.holder |
Copyright: INFORMS |
en |
pubs.author-url |
http://meetings2.informs.org/austin2010/dm-si.html |
en |
pubs.finish-date |
2010-11-06 |
en |
pubs.publication-status |
Accepted |
en |
pubs.start-date |
2010-11-06 |
en |
dc.rights.accessrights |
http://purl.org/eprint/accessRights/RestrictedAccess |
en |
pubs.subtype |
Proceedings |
en |
pubs.elements-id |
197192 |
en |
pubs.org-id |
Engineering |
en |
pubs.org-id |
Engineering Science |
en |
pubs.record-created-at-source-date |
2010-12-15 |
en |