兵棋推演態(tài)勢(shì)認(rèn)知的態(tài)勢(shì)數(shù)學(xué)表達(dá)方法

打開文本圖片集

中圖分類號(hào)：TP391.9 文獻(xiàn)標(biāo)志碼：A DOI：10.12305/j.issn.1001-506X.2025.03.20

situation mathematical expression method of situational awareness in wargame

PAN Chenhui1，*，XIAN Yong1，MA Peiyang1，NI Wancheng2，ZHAO Xiaonan2，LI Shaopeng1，3 （1. Sohoolof Missile Engineerings Rocket Force University of Engineering，Xi'an 710025，China; 2.Instituteof Automation，Chinese Academy of Sciences，Beijing l0ol9o，China； 3.Departmentof Automation，Tsinghua University，Beijing looo84，China）

Abstract：In order to meet the demand for intelligent situationcognition，this paper proposes a standardized construction method of machine learning dataset for batlefield situation cognition in wargame.Based on feature engineering analysis，this study focusesonarmy tactical wargameand analyzes key elements of batlefield situation，and proposes a hierarchical and grid based model for expressng the battlefield situation.By using the equal interval time correlation mode to slide sliceand map transformation the data of the wargame process，the problems of inconsistent military chess data structure，imbalanced feature labels，and diffculty in data transformation fidelityare solved，and the automatic collctionand storage of samples are achieved.A machine learning for situational awarenessin military chess deduction dataset containing 26o Oo samples is constructed and publicly available.The experiment shows thatthedataset constructed using the mapping modelin this paper has fewer subjective colors，good data fidelity，automatic and eficient data colection process，and good dataset consistency.

Keywords：wargame； feature expression； situation awareness；machine learning；data set

0 引言

人工智能技術(shù)的發(fā)展給軍事斗爭(zhēng)準(zhǔn)備帶來巨大沖擊，在信息化戰(zhàn)爭(zhēng)后，智能化戰(zhàn)爭(zhēng)逐步成為各軍事大國(guó)爭(zhēng)相研究的熱點(diǎn)、難點(diǎn)。（剩余17406字）