目标状态拓扑辅助的T/R-R紧凑型地波雷达航迹关联方法

    Target State Topology Assisted Track-to-track Association Method for T/R-R Compact HFSWR

    • 摘要: 针对T/R-R复合双基地紧凑型高频地波雷达在多目标邻近行驶跟踪场景中,由于目标间运动学参数相似导致的航迹错误关联问题,提出了一种目标状态拓扑辅助的航迹关联方法。首先,利用关注目标与其周围船只在空间状态上的一致性,以关注目标为中心,以其周围邻近船只为节点构建拓扑结构;然后,提取关注目标及其邻近船只之间的相对状态关系作为关注目标的空间结构特征,并改进最优子模式分配距离计算结构相似度以辅助航迹关联;最后,结合目标航迹参数相似度和结构相似度,以总关联代价最小作为优化准则,采用Jonker-Volgenant-Castanon算法求解了最优的航迹关联结果。利用仿真与实测数据开展了多目标航迹关联实验,实验结果表明:该方法能够有效解决多目标邻近行驶场景下由于目标运动学参数相似导致的关联模糊问题,提高了航迹关联准确率。

       

      Abstract: To address the issue of track-to-track misassociation caused by similar kinematic parameters among adjacent targets in multi-target nearby tracking scenario for T/R-R compact high-frequency surface wave radar, a target state topology assisted track-to-track association method is proposed in this paper. First, utilizing the spatial state consistency between the target of interest and surrounding vessels, a topological structure is constructed with the target as the center and nearby vessels as nodes. Then, the relative state relationships between the target and nearby vessels are extracted as the spatial structural features of the target and an extended optimal subpattern assignment distance is modified to calculate structural similarity, aiding track-to-track association. Finally, by considering both parametric and structural similarities of target tracks, the optimal association is solved using the Jonker-Volgenant-Castanon algorithm, with the goal of minimizing the total association cost. Experiments on multi-target trajectory association are conducted using simulation and real-measurement data. The results show that the proposed method effectively resolves the ambiguity caused by similar kinematic parameters in multi-target close maneuvering scenarios, improving track association accuracy.

       

    /

    返回文章
    返回