An evolutionary path planning algorithm for military applications

Abstract

The path planning problem for military applications is discussed, with a review of relevant literature. An evolutionary algorithm originally designed for optimizing 3-dimensional highway alignments is adapted and tested for real-time military path planning applications in a changing environment. An optimization problem is formulated to seek a path for an autonomous agent or robot between given origin and destination points. The problempsilas decision variables and constraints are discussed. This problem maximizes the net benefit of reaching the destination while considering the probabilities of destroying hostile targets and getting destroyed by them during the mission. A hypothetical numerical example for a rescue operation at a location in hostile territory is presented. Minimizing time to destination, maximizing survivability and other measures of performance are also discussed. The solution algorithm is intended for real-world path planning for autonomous agents or robots.