Accession Number:

ADA461621

Title:

Increased UAV Task Assignment Performance Through Parallelized Genetic Algorithms (Preprint)

Descriptive Note:

Conference paper

Corporate Author:

INSTITUTE FOR SCIENTIFIC RESEARCH INC FAIRMONT WV

Report Date:

2006-08-01

Pagination or Media Count:

10.0

Abstract:

This paper explores the parallelization of a Genetic Algorithm GA utilized for task assignment of a team of Unmanned Air Vehicles conducting a Suppression of Enemy Air Defense mission. The GA has been developed and implemented in the Multi-UAV simulation environment for testing. The algorithm has been parallelized with each UAV acting as an independent processor. Two different implementations are explored, one where each UAV independently runs a GA, and the best overall solution is selected at the end, and one where the UAVs exchange information several times during the evolution of generations. The results of these implementations are compared to the original, non-parallelized GA performance.

Subject Categories:

  • Pilotless Aircraft
  • Operations Research
  • Cybernetics
  • Military Aircraft Operations
  • Computer Programming and Software

Distribution Statement:

APPROVED FOR PUBLIC RELEASE