Introduction

SARBayes on WordPress

The original SARBayes site died when Hosting Revolution vanished in October 2011 along with their offsite backups.  I recovered what I could and re-launched a Joomla site March 2012.  That went down in June.  I've moved to Bluehost and WordPress.  Wish me better luck.

         -Charles 2012-07-26

Introduction

The SARBayes project and develop models, algorithms, and software to support missing-persons searches on land, using data that we and others have collected about missing-person behavior and search operations. Slowly.  Some projects include:

  • collecting data on Australian searches into an online database
  • making and testing Bayesian network models to predict lost-person behavior
  • developing a general-purpose library for optimal resource allocation
  • (integrating all this search theory into intuitive software)

SARBayes is about search theory, and especially probability maps.  Given a probability map for the current location of the lost person, search theory provides optimal resource allocation algorithms.  But until recently there was no good way to make a probability map for land search. Now there are several approaches, and the MapScore project will provide a way to compare them on actual cases.

The online version of the Australian case data was wiped out with the server.  However, all the cases are now part of ISRID, the International Search & Rescue Incident Database, and we will be using that for our analyses.  We will endeavor to make the Australian cases available again. (ISRID is not.)

Comments are welcome: ctwardy at sarbayes dot org, or ctwardy at alumni dot indiana dot edu

History

SARBayes was started in late 2000 by Charles Twardy during his postdoc in the Reasoning Under Uncertainty Group at Monash University in Melbourne, Australia, with the cooperation of the Victorian Police Search & Rescue Squad, and VicWalk's Bushwalkers' Search & Rescue. (Special thanks to Rob Gatt, Vic Velthuis, and Rik Head.)  From the beginning, the project has received invaluable assistance from other individuals and organizations, especially Bob Koester, Jack Frost, and Alan Washburn.

While at Monash, the Charles worked on SARBayes with David Albrecht, Kevin Korb, Ann Nicholson, and a talented group of students especially Adam Golding and the AGM team (Andre Oboler, Gareth Thompson, and Michael Eldridge).  During this time Charles created the Australian database of lost person behavior and analyzing it.  Adam Golding created the first Bayesian network models of lost person behavior, and a prototype probability mapping program, Probability Mapper.  With Charles and David, the AGM team developed their student project into the SORAL library, and then linked the AGM code, SORAL, and Probability Mapper, albeit loosely.

In 2005, Charles returned to the U.S. on a SBIR grant with Jim Donovan and Bob Koester, and finished the Australian Lost Person Behaviour Report (2006). Those cases helped to prototype the ISRID database.  He was also able to fund AGM to create an updated AGM-SAR that linked tightly with SORAL, and distribute it under the Gnu GPL.

In 2008, Charles joined George Mason University as research faculty, and in 2011 began a collaboration with the WiSAR team at Brigham Young University to develop MapScore, a web portal for scoring and comparing probability maps generated by different teams and methods, using actual cases from ISRID.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.