火力与指挥控制2023，Vol.48Issue(12)：17-26,10.DOI:10.3969/j.issn.1002-0640.2023.12.003
意图驱动的防空武器-目标动态分配方法
Intention Driven Dynamic Weapon-target Allocation Method for Air Defense
摘要
Abstract
Weapon-target allocation is one of the important contents of air defense battle command and decision-making.Most of the studies on this problem have too much consideration of their own situation and limited target information,and the cognition of enemy target intention is rarely involved.Combined with the air target combat intention,an intention driven dynamic weapon-target allocation method for air defense is proposed.(Intent-WTA),Considering the impact of the behaviors of the enemy aircrafts on our decision-making,first of all,for the collected enemy aircraft swarm data,the expert experience is encapsulated as a label,and the two-way long and short term memory network(BiLSTM-Attention)with attention mechanism is trained and fused for intention recognition.Then,according to the intentions of different enemy combat aircrafts,the passive attributes of the enemy aircrafts,the attributes of our weapons,battlefield environment information and real-time dynamic information,the intention driven weapon-target allocation model(Intent-WTA)is proposed.Finally,the cost flow algorithm is used to accurately obtain the optimal weapon-target allocation strategy in a short time.The experimental simulation results show that the proposed method can accurately and efficiently make weapon-target allocation according to the dynamic changes of the battlefield situation,and can meet the practical requirements of air defense operations.关键词
武器-目标分配/意图识别/双向长短期记忆网络/费用流/防空作战Key words
weapon-target allocation/intention recognition/bidirectional long and short term memory network/cost flow/air defense operations分类
计算机与自动化引用本文复制引用
丁鹏,宋亚飞,夏智权..意图驱动的防空武器-目标动态分配方法[J].火力与指挥控制,2023,48(12):17-26,10.基金项目
国家自然科学基金(61806219,61703426,61876189) （61806219,61703426,61876189）
陕西省科学基金(2021JM-226) （2021JM-226）
陕西省高校科协青年人才托举计划(20190108,20220106) （20190108,20220106）
陕西省创新能力支撑计划资助项目(2020KJXX-065) （2020KJXX-065）
