Cooperative path planning for multiple air vehicles using a co-evolutionary algorithm

Abstract

The coordinated path-planning problem for multiple unmanned air vehicles is studied with the proposal of a co-evolving and cooperating path planner. In the new planner, potential paths of each vehicle form their own subpopulation, and evolve only in their own sub-population, while the interaction among all sub-problems is reflected by the definition of fitness function. Meanwhile, the individual candidates are evaluated with respect to the workspace so that the computation of the configuration space is avoided. By using a problem-specific representation of candidate solutions and genetic operators, our algorithm can take into account different kinds of mission constraints and generate the solutions in real-time.