首页|基于RL-LSTM的空中目标意图识别方法

基于RL-LSTM的空中目标意图识别方法

扫码查看
空中目标意图识别是战场态势认知的重要部分。为了进一步提高空中目标意图识别准确率及实时性,提出了基于改进长短时记忆(long short-term memory,LSTM)网络模型RL-LSTM的空中目标意图识别方法。首先获取目标实时的状态数据,以最后时刻目标状态作为模型输入,利用RL-LSTM模型来学习7种常见意图的运动及时间相关特征信息,最后,通过Softmax分类器实现目标意图识别。仿真实验表明,该模型提升了现有神经网络模型的识别准确率及识别效率。
Research on Air Target Intention Recognition Method Based on RL-LSTM
Air target intention recognition is an important part of battlefield situation recognition.In order to further improve the accuracy and real-time performance of air target intention recognition,a method of air target intention recognition based on improved long short term memory(LSTM)network model RL-LSTM is was proposed.Firstly,the real-time state data of the target is obtained,and the last moment state of the target is taken as the model input.The RL-LSTM model is used to learn the motion and time related feature information of seven Kinds of common intentions.Finally,the target intention is recognized by Softmax classifier.The simulation experiments show that this model improves the re-cognition accuracy and efficiency of existing neural network models.

air targetintention recognitionRL-LSTM modelneural network

张鹏程、张勇、李建国、张鹏飞、魏鑫

展开 >

北方自动控制技术研究所,太原 030006

空中目标 意图识别 RL-LSTM模型 神经网络

军委科技委项目

2019XXX

2024

火力与指挥控制
火力与指挥控制研究会,火力与指挥控制专业情报网

火力与指挥控制

CSTPCD北大核心
影响因子:0.312
ISSN:1002-0640
年,卷(期):2024.49(2)
  • 18