| 摘要: |
| 雷达目标跟踪主要利用从回波信号中得到的目标位置、速度等信息，所用的先验信息较少。由于地形、燃料等因素，空中/海面目标通常是会沿着一定航道进行运动，但这一信息通常难以作为先验信息提供给雷达观测，用以辅助雷达跟踪。因此本文提出基于航迹密集程度的区域航道挖掘方法，从航迹数据集提取航道，可为后续相同区域内的目标跟踪提供必要的辅助信息支持。本方法对航迹进行参数拟合，基于参数聚类找到航迹高密集区域并利用最小二乘提取航道；它在无任何先验知识的前提下，仅利用已有航迹数据就可实现对区域内航道快速、准确提取。基于实测数据，通过本方法与其他方法的实验对比，我们验证了本方法提取航迹的有效性和准确性。 |
| 关键词: 机器学习 航道提取 参数拟合 最小二乘 |
| DOI：DOI:10.3969/j.issn.1672-2337.2023.03.014 |
| 分类号:TN953+.6 |
| 基金项目: |
|
| A Density Based Fitting Method for Route Extraction |
|
NIE Yiwen, LIU Junwei
|
|
1. The 38th Research Institute of China Electronics Technology Group Corporation, Hefei 230088, China;2. Key Laboratory of Aperture Array and Space Application, Hefei 230088, China
|
| Abstract: |
| Radar target tracking mainly relies on target locations and velocities extracted from radar echoes, with little priori information. Due to the constraints of terrain and fuel, air/sea targets usually move along certain routes. However, routes are hardly provided to radar as priori information to assist target tracking. Based on this perception, we propose a density based fitting method for route extraction （DFRE） in this paper. It extractes routes from trace set, which can be used to support target tracking. Our method firstly applies parameter fitting on traces, and then find condensed regions of traces based on the clustering of parameters and extract routes by the least square method. DFRE can achieve fast and precise route extraction only by exploiting traces without any priori information of tracking. By experimental comparisons with other state?of?the?art route extraction methods on actual track data, the performance of our method is validated. |
| Key words: machine learning route extraction parameter fitting least square method |
