针对低轨卫星LEO星载GPS实时定轨中存在的问题,提出了以单点定位结果为观测值,采用自适应卡尔曼滤波(AKF)方法进行动力平滑来实现LEO星载GPS实时动力法定轨。采用2004—03—29~31日的GRACE-A卫星实测数据进行了实时定轨计算,并分析了自适应因子、噪声补偿方差、GPS信号中断对自适应定轨的影响。通过计算分析发现,采用AKF进行LEO星载GPS实时定轨可有效解决采用EKF噪声补偿方差难以确定的难题。同时还发现,采用AKF进行LEO星载GPS实时定轨具有较强的稳定性。
To solve the problem of real time orbit determination of LEO satellites based on GPS, the method of dynamic orbit smoothing was put forward using adaptive Kalman filtering(AKF), which taking positioning solutions as observation. The real time of GRACE-A satellite orbit was computed by using the GPS data observed from March 29 to 31,2004. The influence of the adaptive factor, the variance matrix of model noise and the signal of GPS break-off in AKF orbit determination were analyzed. With comparison and analysis, the result showed that the problem of difficult determination of the variance matrix of model noise in extended Kalman filtering(EKF)orbit determination coulld be resolved effectively in AKF orbit determination, and the reliability of AKF orbit determination was strongly in real time orbit determination of LEO satellites based on GPS.