无线电工程2025，Vol.55Issue(10)：2019-2026,8.DOI:10.3969/j.issn.1003-3106.2025.10.008
基于航迹数据图的航迹融合算法
Track Fusion Algorithm Based on Track Data Graph
摘要
Abstract
Multi-source track fusion,which is to provide multi-level target track information with higher accuracy than a single sensor,is one of the research directions in the field of information fusion.Traditional track fusion algorithms have low accuracy and rely on priori states.To solve these problems,a track fusion algorithm based on Track Data Graph(TDG)is proposed.Firstly,the observed trajectories of multiple sensors for the same target are normalized;then the normalized trajectory data are used to build the trajectory data graph;finally,convolutional neural networks are used to complete feature extraction,obtain relational features of target observation trajectories and achieve fused target trajectory.Experimental results show that the track fusion algorithm based on TDG performs optimally in the evaluation of Mean Square Error(MSE),Mean Absolute Error(MAE),and R2 during comparison experiment,and improves the fusion accuracy of target tracks in general.关键词
航迹融合/深度学习/数据图/卷积神经网络Key words
track fusion/deep learning/data graphs/convolutional neural network分类
信息技术与安全科学引用本文复制引用
李祥民,李凯,王元春,郭志文,庞士杰,杨鲜,强保华..基于航迹数据图的航迹融合算法[J].无线电工程,2025,55(10):2019-2026,8.基金项目
桂林市重点研发计划项目(20230110-1)Guilin Key Research and Development Program(20230110-1) （20230110-1）
