Vehicle route planning with constraints using genetic algorithms

Abstract

A route planning approach based on a class of adaptive search techniques called genetic algorithms (GAs) is presented for planning 3D routes for multiple air-vehicles through a threat dense environment. This paper describes a GA-based route planner which generates effective vehicle routes and elegantly accommodates these mission constraints. Preliminary studies on GA-based air-vehicle route planners has shown this approach to be very promising. This paper extends previous research through integration with a complete hierarchy-based mission management system. The results of several experiments are illustrated and discussed. The main thrust of these experiments focus on: (1) investigating effective configuration of classes of GA operators; (2) determining GA operator parameter settings that will produce "near-optimal" routes; (3) exploring the use of a domain-specific mutation operator, called "target bias mutation", for expediting convergence; and (4) comparing results against the well-known dynamic programming algorithm.<>