Multi-criteria shortest path algorithm for global path planning of unmanned combat vehicles

Abstract

We consider a problem of finding paths between two points in a region or terrain for the movement of an unmanned combat vehicle (UCV). To represent the terrain while reflecting a battlefield situation, we use a grid generation method in which to represent the region. The terrain is divided by the grid and each grid is given the path-length costs with the traverse time of the UCV, risk level determined by the enemy position and jamming level influenced by the electronic attack of the enemy. For a problem with a multiple criterion objective of minimizing time, risk and jamming, we develop a two-phase hybrid labeling algorithm with a new label selection strategy and propose a global path planning algorithm based on the label correcting method. For the evaluating of the performance of the proposed algorithm, computational experiments are performed on a number of data sets, and results show that the proposed algorithm performs better than existing methods. We also give results of a case in which the proposed algorithm is applied to a real problem.