复杂对抗下的防空作战中，如何对战场目标的意图进行识别成为防空作战态势认知的关键，针对编队目标意图识别过程中的状态变化灵活、意图规律性强等特点，本文提出了数据-知识双驱动的编队目标意图识别方法，首先设计了数据-知识双驱动的意图识别总体框架。其次基于该框架设计数据驱动的基于双向长短时记忆网络和条件生成对抗网络的目标状态预测（target state prediction based on BiLSTM and CGAN，BC-TSP）方法。而后提出了知识驱动的基于模糊决策树的意图推理方法。最后分别进行实验验证，结果分析表明本方法既有着优秀的时序特征数据学习能力，能准确预测编队目标未来的状态信息，又可以根据实际情况高效利用不完备的先验知识，通过可解释性强的推理方式获得编队目标的最终意图，满足防空作战指挥员的态势认知需求，辅助其更好地进行决策。
In complex adversarial air defense operations, the identification of battlefield target intent has become crucial for situational awareness. To address the characteristics of flexible state variations and strong regularity in the intent of formation targets, this paper proposes a data-knowledge dual-driven method for formation target intent recognition.First, the overall framework of da-ta-knowledge dual-driven intent recognition is designed. Subsequently, based on this framework, a data-driven target state pre-diction (BC-TSP) method is proposed, combining Bidirectional Long Short-Term Memory (BiLSTM) and Conditional Genera-tive Adversarial Network (CGAN). Furthermore, a knowledge-driven intent inference method based on fuzzy decision trees is introduced.Finally, experiments were conducted for validation. The results demonstrate that the proposed method not only exhib-its strong temporal feature learning capability, enabling accurate prediction of future formation target states, but also efficiently leverages incomplete prior knowledge through a highly interpretable reasoning process to determine the final intent of formation targets. This approach satisfies the situational awareness needs of air defense commanders, assisting them in making better-informed tactical decisions.
[1] 张会霞, 梁彦, 马超雄等. 数据和知识驱动的空战目标集群类型综合识别[J/OL]. 航空学报: 1-15[2023-03-29].
[2] 饶佳人, 孙世庆, 李婷婷. 基于规则的空中作战编队识别[J]. 指挥信息系统与技术, 2021, 12(03): 19-22+29.
[3] 张桂林, 吴蔚, 徐建等. 基于规则引擎的海上编队目标识别推理[J]. 指挥信息系统与技术, 2022, 13(03): 28-31+43.
[4] 乔殿峰, 梁彦, 马超雄等. 多域作战下的群目标意图识别与预测[J]. 系统工程与电子技术, 2022, 44(11): 3403-3412.
[5] 杨雨田, 杨杰, 李建国. 基于EMEBN的空中目标战术意图识别方法[J]. 火力与指挥控制, 2022, 47(05): 163-170.
[6] 刘建伟, 谢浩杰, 罗雄麟. 生成对抗网络在各领域应用研究进展[J]. 自动化学报, 2020, 46(12): 2500-2536.
[7] Guanglei M, Runnan Z, Biao W, et al. Target tactical intention recognition in multiaircraft cooperative air combat[J]. International Journal of Aerospace Engi-neering, 2021, 2021: 1-18.
[8] 蒲志强, 易建强, 刘振等. 知识和数据协同驱动的群体智能决策方法研究综述[J]. 自动化学报, 2022, 48(03): 627-643.
[9] Radford A, Metz L, Chintala S. Unsupervised represen-tation learning with deep convolutional generative ad-versarial networks[J]. arXiv preprint arXiv:1511.06434, 2015.
[10] Arjovsky M, Bottou L. Towards principled methods for training generative adversarial networks[J]. arXiv pre-print arXiv:1701.04862, 2017.
[11] 马钰棠, 孙鹏, 张杰勇等. 样本不平衡下的空中群组意图识别方法[J]. 系统工程与电子技术, 2022, 44(12): 3747-3755.
[12] 苟先太, 吴南方. 基于GRU-Attention神经网络的空中群组态势识别方法[J]. 计算机与现代化, 2019(10): 11-16+33.
[13] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[14] 刘钻东, 陈谋, 吴庆宪等. 非完备信息下无人机空战目标意图预测[J]. 中国科学: 信息科学, 2020, 50(05): 704-717.
[15] Mirza M, Osindero S. Conditional Generative Adver-sarial Nets[J]. Computer Science, 2014: 2672-2680.
[16] Zhou T, Chen M, Wang Y, et al. Information entropy-based intention prediction of aerial targets under un-certain and incomplete information[J]. Entropy, 2020, 22(3): 279.
[17] Wang L, Li S. Tactical intention recognition of aerial target based on XGBoost decision tree[J]. Journal of Measurement Science and Instrumentation, 2018, 9(2): 148-152.
[18] 陈浩, 任卿龙, 滑艺等. 基于模糊神经网络的海面目标战术意图识别[J]. 系统工程与电子技术, 2016, 38(08): 1847-1853.
[19] 施伟, 冯旸赫, 程光权等. 基于深度强化学习的多机协同空战方法研究[J]. 自动化学报, 2021, 47(07): 1610-1623.
[20] 刘敏, 郎荣玲, 曹永斌. 随机森林中树的数量[J]. 计算机工程与应用, 2015, 51(05): 126-131.
[21] 强立, 杨凡德, 施令. 空间态势认知产品生成及关键技术研究[J]. 指挥控制与仿真, 2018, 40(03): 5-9+13.
CJA