一种提高GNSS测速精度的自适应Kalman滤波算法

在卫星导航系统动态定位中,采用基于瞬时多普勒观测量的最小二乘法确定速度,当载体高机动时,多普勒误差迅速增大,从而导致测速精度大幅度降低。针对该问题,提出一种同时实现动态模型自适应修正和观测模型自适应更新的Kalman滤波算法。算法采用滑动窗方式来建立实时更新的动态模型参数,使当前统计模型自适应地跟踪载体的动态特性。此外,算法提出观测模型的自适应更新方法,通过设置载体状态判决门限,高、中机动时仅进行受动态应力影响小的伪距更新,低机动下添加精度较高的伪距率更新。通过Sprient GSS8000模拟器产生的动态场景验证表明,相对于最小二乘法和常规Kalman滤波算法,提出的自适应Kalman滤波算法能够全面提高载体在多种运动状态下的测速精度。

An Adaptive Kalman Filtering Algorithm to Improve the Accuracy of GNSS Velocity Determination

To avoid the significant reduction of velocity accuracy brought by the rapid increase of Doppler error for vehicle high maneuver,which exists in the velocity determination using the Least Square method based on Doppler observations of GNSS kinematic positioning,this paper presents a Kalman filtering algorithm implementing both dynamic model and measurement model adaptive adjustment.In the algorithm,a sliding window is used to update the model parameters in real time,making the current statistical model matc...