摘要
装备的认知能力集中体现了认知电子战的根本属性,也是能否对复杂、智能的电子装备实施有效干扰的关键。强化学习作为人工智能领域炙手可热的技术手段,具备不依赖先验数据的自学习能力,是解决多功能雷达干扰难题的一个重要途径。在回顾传统的雷达干扰决策算法的基础上,阐述了利用强化学习进行雷达干扰决策的原理,分析了基于强化学习的雷达干扰决策技术的发展现状,并通过仿真分析了基于强化学习的干扰决策方法的性能,最后进行了总结与展望。
The cognitive ability of equipment embodies the fundamental attributes of cognitive electronic warfare,and it is also the key to effective jamming on complex and intelligent electronic equipment.As a hot technique in the field of artificial intelligence,reinforcement learning has the ability of self-learning not relying on priori data,which is an important approach to multifunctional radar jamming.Based on the review on traditional algorithms of radar jamming decision-making,as for reinforcement learning based radar jamming decision-making,the principles and status quo of the technology are analyzed,and its performance is verified through simulations.Finally,a summary is given,and the outlook for the technology is predicted.
作者
朱霸坤
朱卫纲
李伟
杨莹
高天昊
ZHU Bakun;ZHU Weigang;LI Wei;YANG Ying;GAO Tianhao(University of Aerospace Engineering Graduate School of Aerospace Engineering,Beijing 101000,China;University of Aerospace Engineering Department of Electronics and Optics,Beijing 101000,China)
出处
《电光与控制》
CSCD
北大核心
2022年第4期52-58,111,共8页
Electronics Optics & Control
基金
复杂电磁环境效应国家重点实验室项目(2020Z0203B)。