基于经纬度的航迹数据改进抗野值Kalman滤波

摘要

Due to adverse impact of the track data outliers on the test in real-time target tracking,and the lack of X,Y,Z ax-is speed,the paper proposes for the first time based on the longitude and latitude to remove outliers.Three correction methods of re-al-time anti field values are studied,the threshold method,the least square method and the anti-outlier Kalman filtering method. In view of its deficiency,the improved algorithm of Kalman filter is proposed.According to the threshold method to determine the initial value,improved Kalman gain update strategy,by extrapolation and compensation point for outliers,improve the outlier recog-nition rate to accurately real-time grasp the target information.The simulation test and flight test show that the method is stable and effective,and it can eliminate the isolated and contiguous outliers effectively.It has been applied in the test task.%由于实时目标跟踪中航迹数据中的野值对试验造成不利影响,且数据缺乏X、Y、Z轴的速度,为此,文章首次提出基于经度、纬度进行野值剔除.研究了三种实时抗野值的修正方法:简单经纬度法、最小二乘法和抗野值Kalman滤波方法.针对其不足,提出改进抗野值Kalman滤波算法,根据改进简单经纬度法确定初始真值,改进Kalman增益更新策略,通过外推和野值补点,提高野值识别率,能够较准确地实时掌握目标信息.经仿真试验和飞行试验验证,该方法稳定有效,对单点野值和成片野值均能有效剔除,且已应用在试验任务中.