1.西北工业大学 航天学院, 西安 710072
2.中国运载火箭技术研究院 研究发展中心, 北京 100006
黄虎(1986-), 男, 硕士, 研究员。 E-mail: huanghu7@126.com
收稿:2026-04-02,
修回:2026-07-27,
网络首发:2026-08-17,
移动端阅览
李猛,果昊涵,黄虎,等. 基于生成式知识迁移的多飞行器多目标任务分配[J]. 航空工程进展.
LI Meng,GUO Haohan,HUANG Hu, et al. Multi-vehicle multi-target task assignment based on generative knowledge transfer[J]. Advances in Aeronautical Science and Engineering.(in Chinese)
多飞行器多目标任务分配存在环境不确定性和任务高动态变化性,导致传统算法分配结果的有效性和实时性难以保障。本文考虑飞行器位置、速度信息以及目标价值信息,构建多飞行器多目标任务分配模型,研究基于生成式知识迁移的任务分配算法,引入生成式网络生成分配方案,通过Gumbel-Softmax实现可微化处理,解决离散分配方案难以梯度反向传播的问题;为说明本文分配算法的有效性和实时性,与传统智能算法在静态和动态两类场景进行对比仿真验证。结果表明:分配算法提升了任务分配有效性和实时性,可为多飞行器多目标任务分配方法提供参考。
Aiming at the problems of environmental uncertainty and high dynamic task variation in multi-vehicle multi-target assignment, traditional algorithms struggle to guarantee the effectiveness and real-time performance of allocation results. This paper constructs a multi-vehicle multi-target assignment model by considering the position and velocity information of vehicles as well as target value information. A task allocation algorithm based on generative knowledge transfer is studied: a generative network is introduced to generate allocation schemes, and the Gumbel-Softmax technique is adopted to achieve differentiable processing for discrete task selection, thereby solving the problem that discrete assignment schemes are not amenable to gradient backpropagation. To verify the effectiveness of the proposed algorithm, comparative simulations with traditional intelligent algorithms are conducted in both static and dynamic scenarios. The results show that the proposed algorithm can effectively improve the real-time performance and effectiveness of assignment, which can provide a valuable reference for the research and development of multi-vehicle multi-target assignment in high-dynamic scenarios.
Rodriguez C A , Jenkins P R , Robbins M J . Solving the joint military medical evacuation problem via a random forest approximate dynamic programming approach [J]. Expert Systems With Applications , 2023 , 221 : 1 - 11 .
Maheswari U A , Suganthi P . Designing energy-efficient military communication using modified intelligent water drops algorithm on MANETs [J]. International Journal of Information Technology , 2023 , 15 ( 2 ): 991 - 1003 .
徐庆征 , 马依笑 , 王娜 . 不确定环境下多无人机任务分配 [J]. 海军航空大学学报 , 2023 , 38 ( 6 ): 473 - 482 .
Xu Qingzheng , Ma Yixiao , Wang Na . Task Allocation of Multiple UAVs in Uncertain Environment [J]. Journal of Naval Aviation University , 2023 , 38 ( 6 ): 473 - 482 . (in Chinese)
刘诗瑶 , 王明 , 张耀 , 等 . 基于深度强化学习的无人战车多回合火力分配 [J]. 火力与指挥控制 , 2023 , 48 ( 7 ): 79 - 84 .
Liu Shiyao , Wang Ming , Zhang Yao , et al . Multi-round Firepower Allocation for Unmanned Combat Vehicles Based on Deep Reinforcement Learning [J]. Fire Control & Command Control , 2023 , 48 ( 7 ): 79 - 84 . (in Chinese)
赵国亮 , 穆培培 , 李泽文 . 求解静态武器—目标分配问题的混合优化算法 [C]∥ 第十三届中国指挥控制大会 . 北京 : 中国指挥与控制学会 , 2025 : 74 - 81 .
Zhao Guoliang , Mu Peipei , Li Zewen . Hybrid Optimization Algorithm for Solving Static Weapon-Target Assignment Problem [C]∥ The 13th China Command and Control Conference . Beijing : Chinese Institute of Command and Control Beijing , 2025 : 74 - 81 . (in Chinese)
任涛 , 吴晶 , 易亮 , 等 . 武器-目标分配问题的智能优化算法研究进展 [J]. 计算机仿真 , 2024 , 41 ( 12 ): 1 - 6, 12 .
Ren Tao , Wu Jing , Yi Liang , et al . Research Progress on Intelligent Optimization Algorithms for Weapon-Target Assignment Problem [J]. Computer Simulation , 2024 , 41 ( 12 ): 1 - 6, 12 . (in Chinese)
王光源 , 李浩民 , 陈永展 , 等 . 基于灰狼算法的多平台火力分配优化模型 [J]. 科学技术与工程 , 2023 , 23 ( 23 ): 10133 - 10138 .
Wang Guangyuan , Li Haomin , Chen Yongzhan , et al . Optimization model of multi-platform firepower allocation based on grey wolf algorithm [J]. Science Technology and Engineering , 2023 , 23 ( 23 ): 10133 - 10138 . (in Chinese)
褚凯轩 , 常天庆 , 张雷 . 基于改进人工蜂群算法的地面作战武器-目标分配 [J]. 兵工学报 , 2023 , 44 ( 7 ): 2171 - 2183 .
Chu Kaixuan , Chang Tianqing , Zhang Lei . Ground combat weapon-target allocation based on improved artificial bee colony algorithm [J]. Acta Armamentarii , 2023 , 44 ( 7 ): 2171 - 2183 . (in Chinese)
张杰 , 吴曦 , 李元龙 . 基于改进果蝇优化算法的混合火力分配方法 [J]. 火力与指挥控制 , 2023 , 48 ( 2 ): 86 - 89, 96 .
Zhang Jie , Wu Xi , Li Yuanlong . Hybrid firepower allocation method based on improved fruit fly optimization algorithm [J]. Fire Control & Command Control , 2023 , 48 ( 2 ): 86 - 89, 96 . (in Chinese)
傅勉 , 王丹丹 . 使用贪心模拟退火算法求解WTA问题 [J]. 河北北方学院学报(自然科学版) , 2020 , 36 ( 1 ): 9 - 12, 36 .
Fu Mian , Wang Dandan . Solving the WTA problem using a greedy simulated annealing algorithm [J]. Journal of Hebei North University (Natural Science Edition) , 2020 , 36 ( 1 ): 9 - 12, 36 . (in Chinese)
常雪凝 , 石建迈 , 陈超 , 等 . 基于匈牙利-模拟退火算法的多阶段武器目标分配方法 [J]. 系统工程与电子技术 , 2023 , 45 ( 11 ): 3516 - 3523 .
Chang Xuening , Shi Jianmai , Chen Chao , et al . Multi-stage weapon target allocation method based on Hungarian-simulated annealing algorithm [J]. Systems Engineering and Electronics , 2023 , 45 ( 11 ): 3516 - 3523 . (in Chinese)
李世旺 , 杜溢华 , 胡珊 , 等 . 基于改进量子遗传算法的集群空战多目标分配研究 [C]∥ 第十三届中国指挥控制大会 . 北京 : 中国指挥与控制学会 , 2025 : 274 - 280 .
Li Shiwang , Du Yihua , Hu Shan , et al . Research on multi-target allocation in cluster air combat based on improved quantum genetic algorithm [C]∥ The 13th China Command and Control Conference . Beijing : Chinese Institute of Command and Control , 2025 : 274 - 280 . (in Chinese)
翁年凤 , 刘艺 , 郑奇斌 , 等 . 基于遗传蚁群算法的武器目标分配优化方法 [J]. 海军航空大学学报 , 2024 , 39 ( 5 ): 640 - 648 .
Weng Nianfeng , Liu Yi , Zheng Qibin , et al . Optimization method for weapon-target allocation based on genetic-ant colony algorithm [J]. Journal of Naval Aviation University , 2024 , 39 ( 5 ): 640 - 648 . (in Chinese)
邱少明 , 刘良玉 , 黄昕晨 , 等 . 基于改进PSO-DQN的动态火力分配算法 [J]. 电光与控制 , 2025 , 32 ( 6 ): 24 - 30 .
Qiu Shaoming , Liu Liangyu , Huang Xinchen , et al . Dynamic firepower allocation method based on improved PSO-DQN [J]. Electronics Optics & Control , 2025 , 32 ( 6 ): 24 - 30 . (in Chinese)
李洋 , 刘耿 , 胡晓惠 , 等 . 基于SDP的动态武器目标分配决策算法研究 [J]. 航空兵器 , 2023 , 30 ( 5 ): 50 - 56 .
Li Yang , Liu Geng , Hu Xiaohui , et al . Research on dynamic weapon-target assignment decision-making algorithm based on semidefinite programming (SDP) [J]. Aero Weaponry , 2023 , 30 ( 5 ): 50 - 56 . (in Chinese)
Sonuc E , Sen B , Bayir S . A parallel simulated annealing algorithm for weapon-target assignment problem [J]. International Journal of Advanced Computer Science and Applications , 2017 , 8 ( 4 ): 1 - 8 .
Durgut R , Kutucu H , Akleylek S . An artificial bee colony algorithm for solving the weapon target assignment problem [R]. Samsun : Ondokuz Mayıs University , 2017 .
Zheng G , Gang G , Han J G . Research on autonomous decision method of multi-aircraft cooperative mid-range confrontation [C]∥ Seventh Symposium on Novel Photoelectronic Detection Technology and Applications . Reston : AIAA , 2021 : 11763 .
Li W , Lyu Y , Dai S , et al . A multi-target consensus-based auction algorithm for distributed target assignment in cooperative beyond-visual-range air combat [J]. Aerospace , 2022 , 9 : 486 .
Goodfellow I , Pouget-Abadie J , Mirza M , et al . Generative Adversarial Nets [M]. New York : MIT Press , 2014 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010602201714号