摘要
针对空战环境下多目标轨迹预测问题,提出了基于航迹匹配与麻雀搜索算法(SSA:Sparrow Search Algorithm)-门控循环单元(GRU:Gate Recurrent Unit)的多目标轨迹预测方法。首先基于归属未知的多目标航迹点时序数据,采用基于密度的带噪声空间聚类算法(DBSCAN:Density-Based Spatial Clustering of Applications with Noise)实现在目标个数未知的情况下多目标-航迹匹配;在此基础上,构建目标轨迹预测GRU模型,并采用SSA对GRU网络参数进行优化,确定最优神经元个数,提升轨迹预测模型性能。仿真结果表明,基于航迹匹配与SSA-GRU的多目标轨迹预测方法能够有效识别匹配多个目标航迹,并实现各目标轨迹预测。
In order to address the issue of multi-target trajectory prediction in an air combat environment,a method for multi-target trajectory prediction based on track matching and the Sparrow Search Algorithm(SSA)-Gate Recurrent Unit(GRU)is proposed.Firstly,based on the time series data of multi-target track points with unknown attribution,the Density-Based Spatial Clustering of Applications with Noise(DBSCAN)algorithm is used to achieve multi-target track matching when the number of targets is unknown.On this basis,a target trajectory prediction GRU model is constructed,and SSA is used to optimize the GRU network parameters,determine the optimal number of neurons,and enhance the performance of the trajectory prediction model.The simulation results demonstrate that the multi-target trajectory prediction methodbased on track matching and SSA-GRU can effectively identify and match multiple target paths,enabling the trajectory prediction for each target.
作者
周同乐
ZHOU Tong-le(College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)
出处
《航空电子技术》
2024年第2期40-47,共8页
Avionics Technology
基金
国家自然科学基金青年科学基金(62203217)
航空电子综合与体系集成全国重点实验室基金
江苏省基础研究计划自然科学基金青年基金(BK20220885)。
关键词
目标-航迹匹配
DBSCAN
GRU
麻雀搜索算法
多目标轨迹预测
target-track matching
density-based spatial clustering of applications with noise
gate recurrent unit
sparrow search algorithm
multi-target trajectory prediction
