A Cooperative Task Assignment Method of Multi-UAV Based on Self Organizing Map

摘要

Cooperative task assignment of multi unmanned aerial vehicles (UAV) is an important assurance for UAV to complete military missions. The current task assignment methods have drawbacks such as long execution time and failure to consider UAV's own capacity constraints. Therefore in this paper, we comprehensively considers the total flight distance of the fleet and the weapon load of each UAV, models the cooperative task assignment problem of multi-UAV as a capacitated vehicle routing problem (CVRP). A method based on a competitive neural network called self-organizing feature map (SOM) is used to solve the task assignment problem. The performance of proposed method is compared to that of ant colony algorithm. Experimental results show the feasibility and efficiency of the method.