多平台数据融合系统的航迹关联技术

对于多平台数据融合模式下的航迹关联问题,使用了聚类关联的方法进行解决.采用基于地心坐标系的最小二乘方法对航迹数据进行配准,对航迹间的距离使用Hausdorff距离进行衡量.使用了K-均值算法对各平台侦测的航迹进行关联,并将初始聚类中心设定为相距最远的航迹,有效降低了经典K-均值算法过于依赖初始聚类点带来的错误.仿真数据证实,能在目标密度大且航迹存在交错的场景下保持较高的关联正确率,具有较好的可用性.

Research on track association technology of multi-platform data fusion system

For the track association problem in multi-platform data fusion mode, clustering association method is used to solve it. The least square method based on the geocentric coordinate system (ECEF) is used to register the track data. The distance between tracks is measured by Hausdorff distance. The K-means method is used to correlate the tracks detected by each platform. The initial clustering center is set as the farthest track, which effectively reduces the error caused by the classical k-means algorithm relying too much on the initial clustering points. The simulation data shows that it can keep high correlation accuracy and good usability in the scene of high target density and staggered tracks.