Tactical Maneuver Strategy Learning from Land Wargame Replay Based on Convolutional Neural Network

doi:10.16182/j.issn1004731x.joss.21-0429

Abstract

Abstract:

Aiming at collecting the high valuable knowledge of action decisions in "man-in-the-loop" wargame's replay data, a method of using convolutional neural network to learn the tactical maneuver strategy model from the replay data of wargame is proposed. In this method, the tactical maneuver strategy is modeled as a classification problem of making a good choice from the target candidate locations under the influence of current situation. The key factors affecting commander's decision-making are summarized, and the basic situation features are defined, which are composed of seven attributes such as "maneuverability range and observation range". The feature dataset with positive and negative labels is established. The classifier based on convolutional neural network is designed, which can predict the maneuver terminal position of a single piece by the classification probability. Experimental results show that the prediction accuracy of the tactical maneuver strategy model based on the convolutional neural network is up to 78.96%, which is improved by at least 4.59% compared with other models.

Key words: wargame, replay data, tactical maneuver strategy, situation feature, convolutional neural network

CLC Number:

TP391.9

Jiale Xu, Haidong Zhang, Donghai Zhao, Wancheng Ni. Tactical Maneuver Strategy Learning from Land Wargame Replay Based on Convolutional Neural Network[J]. Journal of System Simulation, 2022, 34(10): 2181-2193.

Figures/Tables 20

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Fig. 11

Table 1

Table 2

Table 3

Table 4

Fig. 12

Table 5

Table 6

Table 7

Fig. 13

References 16

1	胡晓峰, 荣明. 智能化作战研究值得关注的几个问题[J]. 指挥与控制学报, 2018, 4(3): 195-200.
	Hu Xiaofeng, Rong Ming. Several Important Questions of Intelligent Warfare Research[J]. Journal of Command and Control, 2018, 4(3): 195-200.
2	邢思远, 倪晚成, 张海东, 等. 基于兵棋复盘数据的武器效用挖掘[J]. 指挥与控制学报, 2020, 6(2): 132-140.
	Xing Siyuan, Ni Wancheng, Zhang Haidong, et al. Mining of Weapon Utility Based on the Replay Data of War-Game[J]. Journal of Command and Control, 2020, 6(2): 132-140.
3	张俊恒. 计算机兵棋中兵力机动路径规划研究[D]. 长沙: 国防科学技术大学, 2010.
	Zhang Junheng. Research on Force Maneuvering Path Planning in Computer Warfare[D]. Changsha: National University of Defense Technology, 2010.
4	胡伟. 计算机兵棋中兵力机动路径优化研究[D]. 长沙:国防科学技术大学, 2010.
	Hu Wei. The Research on Route Optimization of Military Power Maneuvering in Computer Wargame[D]. Changsha: National University of Defense Technology, 2010.
5	周小镜. 基于改进A*算法的游戏地图寻径的研究[D].重庆: 西南大学, 2011.
	Zhou Xiaojing. Research of Routing in the Game Map Based on Improved A* Algorithm[D]. Chongqing: Southwest University, 2011.
6	刘满, 张宏军, 郝文宁, 等. 战术级兵棋实体作战行动智能决策方法[J]. 控制与决策, 2020, 35(12): 2977-2985.
	Liu Man, Zhang Hongjun, Hao Wenning, et al. Research on Intelligent Decision-making Method of Tactical-Level Wargames[J]. Control and Decision, 2020, 35(12): 2977-2985.
7	朱丰, 胡晓峰, 吴琳, 等. 从态势认知走向态势智能认知[J]. 系统仿真学报, 2018, 30(3): 761-771.
	Zhu Feng, Hu Xiaofeng, Wu Lin, et al. From Situation Cognition Stepped into Situation Intelligent Cognition[J]. Journal of System Simulation, 2018, 30(3): 761-771.
8	闫科, 蔡亚. 陆军合同战术兵棋推演[M]. 北京: 军事科学出版社, 2013: 6-7.
	Yan Ke, Cai Ya. Army Contract Tactical War Game [M]. Beijing: Military Science Press, 2013: 6-7.
9	潘毅. 人机对抗中位置估计及其应用[D]. 北京: 中国科学院大学, 2018.
	Pan Yi. Research on the Location Estimation for Human-Computer System and Its Application[D]. Beijing: University of Chinese Academy of Sciences, 2018.
10	王桂起, 刘辉, 朱宁. 兵棋技术综述[J]. 兵工自动化, 2012, 31(8): 38-41, 45.
	Wang Guiqi, Liu Hui, Zhu Ning. A Survey of War Games Technology[J]. Ordnance Industry Automation, 2012, 31(8): 38-41, 45.
11	刘海洋, 唐宇波, 胡晓峰, 等. 基于兵棋推演的联合作战方案评估框架研究[J]. 系统仿真学报, 2018, 30(11): 4115-4122, 4131.
	Liu Haiyang, Tang Yubo, Hu Xiaofeng, et al. Research on Evaluation Framework of COA Based on Wargaming[J]. Journal of System Simulation, 2018, 30(11): 4115-4122, 4131.
12	徐佳乐. 陆战兵棋态势特征数据集[EB/OL][2021-07-18]. , 2021.
	Xu Jiale. Situation Feature Dataset for Land Wargame[EB/OL][2021-07-18]. , 2021.
13	Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-scale Image Recognition[EB/OL]. [2021-07-18]. .
14	He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.
15	He K, Zhang X, Ren S, et al. Identity Mappings in Deep Residual Networks[C] //European Conference on Computer Vision. Amsterdam: Springer, Cham, 2016: 630-645.
16	Ioffe S, Szegedy C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift[C]//International conference on Machine Learning. Lille: PMLR, 2015: 448-456.

棋子所属方	RED	RED
回合起始位置	90 016	140 042
回合终点位置	100 026	150 043
样本标签	1	1
地图编号	83	86
主要夺控点位置	80 048	150 049
次要夺控点位置	100 049	130 046

地图编号	样本数量
83	10 024
86	6 670
82	3 920
19	3 414
84	52

模型	准确率/%
LR	67.48
SVM	69.08
Resnet	72.98
PCA+SVM	73.15
KNN	74.29
VGG	74.37
CNN	78.96

卷积层个数	测试准确率/%
1	78.21
2	78.61
3	78.75
4	78.53
5	77.72

FC1神经元个数	FC2神经元个数	测试准确率/%
1 024	256	78.24
2 048	256	78.57
4 096	256	78.90
8 192	256	78.57
4 096	512	78.88
4 096	1 024	78.76
4 096	2 048	78.75