摘要
导弹反舰作战是一种有效的非对称打击作战样式,目标毁伤预测在其中承担着关键一环。针对现有反舰导弹目标毁伤预测方法在计算代价或样本规模等方面的限制,在贝叶斯网络技术的框架下,提出一种用于目标毁伤预测的贝叶斯推断模型。该模型将导弹反舰作战的突防和毁伤等过程抽象成高层的概率语义,仅需少量样本数据就可驱动模型关键参数的学习过程,在实现目标毁伤预测的同时还附加有发射弹量规划能力。通过实验,验证了模型的可行性与有效性。
Anti-ship combat with missiles is an effective asymmetric strike combat style,in which target damage prediction plays a key role. Regarding the restrictions of the current target damage prediction methods with anti-ship missiles in computational cost,or sample size and so on,a Bayesian inference model for target damage prediction was built in the framework of Bayesian network technology.This model abstracts the penetration and damage and other processes of missile anti-ship combat into high-level probabilistic semantics,it requires only a small amount of sample data to drive the learning process for the key parameters of the model,and it is able to make target damage while making fired rounds planning. The experimental result also verified the feasibility and validity of the model.
作者
李聪
关爱杰
陈健
徐博奥
LI Cong;GUAN Ai-jie;CHEN Jian;XU Bo-ao(Unit 96901 of PLA,Beijing 100094,China)
出处
《火力与指挥控制》
CSCD
北大核心
2021年第5期98-102,共5页
Fire Control & Command Control
关键词
反舰作战
目标毁伤预测
贝叶斯网络
模型推断
anti-ship combat
target damage prediction
bayesian network
model inference
