体系作战下巡航导弹的动态隐身  

Dynamic stealth of cruise missile in system combat

作　　者：马子杰 谢拥军[1] MA Zijie;XIE Yongjun(School of Electronic and Information Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学电子信息工程学院,北京100191

出　　处：《系统工程与电子技术》2022年第9期2826-2831,共6页Systems Engineering and Electronics

基　　金：国家自然科学基金(61571022,61971022);国家重点实验室基金(HTKJ2019KL504013,61424020305)资助课题。

摘　　要：通过体系作战下的系统作战仿真平台,利用强化学习方法实时调整巡航导弹航迹,实现面对敌方传感器的动态隐身。以巡航导弹突防美军“海军综合防空火控”系统为典型作战场景,基于体系作战仿真平台设计了一种引入动态预警威胁和拦截脱靶量的奖励函数,使用深度确定性策略梯度算法训练巡航导弹突防模型,模型训练完成后巡航导弹通过感知预警机和拦截导弹的状态,改变其机动轨迹以降低探测概率和提高拦截脱靶量,达到动态隐身的效果,进而提升其战场生存能力和突防能力。Through the system combat simulation platform under system combat,this paper uses the reinforcement learning method to adjust the cruise missile trajectory in mission to achieve dynamic stealth facing the enemy’s sensors.Taking cruise missiles to penetrate the US military’s“Navy Integrated Fire Control and Control Air”system as a typical combat scenario,based on the system combat simulation platform,a reward function is designed by including dynamic early warning threats and intercept misses,and the deep deterministic policy gradient algorithm is used to train the cruise missile penetration model.After the model training is completed,cruise missiles can perceive the state of early warning aircraft and intercepting missiles,and change their maneuvering trajectories to reduce detection probability and increase interception misses to achieve dynamic stealth effect,and then enhance its battlefield survivability and penetration ability.

关 键 词：巡航导弹 深度确定性策略梯度算法 雷达隐身 突防策略 

分 类 号：TJ76[兵器科学与技术—武器系统与运用工程]