Exploration a two sided rendezvous search problem using genetic algorithms.
by
Tracy Truong
Aeronautical and Maritime Research Laboratory, DSTO
Coauthors: Andrew Stacey (Royal Melbourne Institute of Technology)

A lost walker being searched for by a helicopter is a two sided search problem when the walker (the target) uses information about the location of the helicopter to modify their behaviour. The target has a smaller speed than the helicopter, but can detect the helicopter at a greater range than the helicopter can detect them. In our case the walker will move towards the helicopter when they can hear it otherwise they will remain stationary. This problem has been explored for a limited set of NATO search paths being scan search, expanding square and decreasing square by Thomas et al. This paper uses a genetic algorithm to search over the space of all possible search paths to find ones which are optimal or near optimal for a range of target speeds and search region sizes. The genetic algorithm was developed using a stationary target as a test case where the optimal solution is known. Details of the genetic algorithm and it performance on this problem will also be presented.

Thomas LC and Hulme PB "Searching for targets who want to be found", Journal of the Operational Research Society 1997 vol. 48 pp44-50.

Date received: July 3, 2001

Copyright © 2001 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cahp-49.