Cooperative Occupancy Decision Making of Multi-UAV in Beyond-Visual-Range Air Combat: A Game Theory Approach

• Published in: IEEE access Vol. 8; pp. 11624 - 11634
• Main Authors: Ma, Yingying, Wang, Guoqiang, Hu, Xiaoxuan, Luo, He, Lei, Xing
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Air combat; Algorithms; Atmospheric modeling; BVR air combat; Combat aircraft; Computing time; Decision making; Decision theory; double oracle algorithm; Game theory; Games; Mathematical model; Missiles; Multiple UAVs cooperative occupancy decision making; multiple UAVs cooperative target assignment; Occupancy; Unmanned aerial vehicles; Zero sum games; zero-sum matrix game
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2019.2933022

In this paper, a new cooperative occupancy decision-making problem for multiple unmanned aerial vehicles (UAVs) in beyond-visual-range (BVR) air combat is proposed; that is, UAV formation on each side first decides the occupancy positions of their own UAVs, then decides the target that each UAV will confront, so that each side can occupy the greatest predominance and be subject to the smallest threat possible. By analyzing the influence of the occupancy positions of the UAVs on the predominance gained by both sides and the threat against them, the problem is described as a zero-sum matrix game. Because the strategy combinations of this game are very numerous, as the best algorithm for the zero-sum matrix game, the Double Oracle (DO) algorithm still has difficulty solving the game model effectively. Therefore, we propose a DO combined with neighborhood search (DO-NS) algorithm to deal with the large-scale zero-sum matrix game. The results show that compared with the DO algorithm, the DO-NS algorithm can ensure the effectiveness of the computational time and solution quality.