系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (12): 3756-3765.doi: 10.12305/j.issn.1001-506X.2022.12.20

• 系统工程 • 上一篇    

基于遗传模糊树的海空对抗无人机智能决策模型

王玉佳1, 方伟1,*, 徐涛1, 余应福1, 邓博元2   

  1. 1. 海军航空大学航空作战勤务学院, 山东 烟台 264001
    2. 海军航空大学岸防兵学院, 山东 烟台 264001
  • 收稿日期:2021-05-07 出版日期:2022-11-14 发布日期:2022-11-24
  • 通讯作者: 方伟
  • 作者简介:王玉佳 (1994—), 女, 硕士研究生, 主要研究方向为智能决策、人工智能|方伟(1977—), 男, 教授, 博士, 主要研究方向为航空兵装备仿真、计算机生成兵力|徐涛 (1985—), 男, 副教授, 博士, 主要研究方向为航空兵装备仿真、计算机生成兵力|余应福 (1984—), 男, 讲师, 硕士, 主要研究方向为航空兵装备仿真、计算机生成兵力|邓博元 (1997—), 男, 硕士研究生, 主要研究方向为人工智能图像处理

Intelligent decision-making model by unmanned aerial vehicles in sea-to-air confrontation based on genetic fuzzy trees

Yujia WANG1, Wei FANG1,*, Tao XU1, Yingfu YU1, Boyuan DENG2   

  1. 1. School of Aviation Operations and Support, Naval Aviation University, Yantai 264001, China
    2. Coastal Defense College, Naval Aviation University, Yantai 264001, China
  • Received:2021-05-07 Online:2022-11-14 Published:2022-11-24
  • Contact: Wei FANG

摘要:

随着人工智能技术的应用领域不断扩大, 无人机(unmanned aerial vehicle, UAV)的自主智能化发展成为目前军事领域的研究热点。其中, UAV的武器智能决策能力是其面临舰艇编队剧烈火力打击时能否顺利完成对海侦察任务的关键条件。针对这一军事应用难点, 利用改进的遗传模糊树构建UAV武器智能决策模型, 以此实现UAV的智能武器决策: 通过设计三模糊子集参数编码法改进参数编码方式, 克服了无效编码导致的系统紊乱; 采用单、组合场景结合的训练方法改进种群训练方式, 提高了决策模型在多样化实战中的适应性。仿真结果表明,所提武器智能决策模型具有较好的有效性和灵活性。

关键词: 遗传模糊树, 武器智能决策, 无人机, 海空对抗

Abstract:

Owing to the continuous expansion of artificial intelligence technology, the development of autonomous intelligence of unmanned aerial vehicle (UAV) has become a research hotspot in the military field. Intelligent weapon decision-making ability is essential for UAV to successfully complete a reconnaissance task in the face of severe fire attack from warships. In view of this difficulty in military application, the improved genetic fuzzy tree is used to build a UAV intelligent weapon decision model, so as to realize the intelligent weapon decision of UAV. The parameter encoding method is improved by designing a parameter encoding method for three fuzzy subsets, thereby overcoming the system disorder caused by invalid encoding. The training method of combining single and combined scenes is used to improve the population training mode, which subsequently improves the adaptability of the decision-making model in diversified actual combat. The simulation results show that the proposed intelligent decision model of weapons has good effectiveness and flexibility.

Key words: genetic fuzzy tree, intelligent weapon decision-making, unmanned aerial vehicle (UAV), sea-to-air confrontation

中图分类号: