Multi-UAV Cooperative Target Assignment Based on Improved NSGA-III Algorithm

Abstract

The weapon-target assignment problem is the key to the combat mission of the UAV against the enemy in the battlefield environment. The purpose is to find a reasonable weapon target assignment scheme based on the threat, value and damage probability of the target, so as to improve the combat efficiency. Aiming at the problem that the current multi-objective optimization algorithm has slow convergence speed and poor convergence stability when solving the static weapon-target assignment problem, and it is difficult to adapt to the high real-time performance of the current battlefield, an improved non-dominated sorting genetic algorithm based on reference points is proposed. The initial population is optimized by binary coding attack scheme, and adaptive mutation and crossover strategy as well as population optimization update strategy is introduced. Based on the threat matrix and advantage matrix obtained by evaluating the battlefield situation, the target attack scheme is generated after multiple iterations of the population. Finally, the Pareto solution set satisfying the constraint condition is calculated, and the relative optimal solution in the Pareto frontier is taken as the attack scheme of multi-UAV. Multiple experiments show that under good conditions, the improved algorithm reduces convergence time by 46.74%, reduces target threat value by 50.5%, reduces total flight range by 26.46%, and increases the number of killing targets by 11.76% compared with the original algorithm. It is proved that the algorithm is reasonable and efficient in solving the problem of target assignment of multi-UAV air-to-ground strike mission.

Keywords

• |air-to-ground attack|multi-uav|weapon target assignment|multi-objective optimization|nsga-iii|pareto solution