提高GPS定位精度的改进卡尔曼滤波算法研究

Study of Improved Arithmetic of Kalman Filter on How to Improve the Precision with Global Position System

介绍了一种有效提高GPS定位精度的改进卡尔曼滤波算法。该算法针对最小二乘法和标准卡尔曼滤波的特点,通过伪距估计出接收机的位置和钟差,有效避免了由于滤波初值、系统噪声方差以及量测噪声方差带来的滤波发散问题。同时该算法不直接使用卡尔曼滤波来估计接收机的状态,而是估计接收机状态的误差,减小了运算量,有效提高了定位精度。在进行状态误差估计时,不需要存储大量测量数据,能方便地进行动态测量数据的实时处理。仿真结果证明此算法具有较快的收敛速度和较高的定位精度。

In this paper, we introduce a new algorithm of improving GPS positioning precision. This algorithm is based on the least - square method and normal Kalman filter. This algorithm can effectively restrain the filter＇s divergence caused by the initial value of filter and system - noise - variance and measurement - noise - variance. The receiver position and clock offset can be estimated via measuring the pseudo ranges. Kalman filter is used to confirm the errors of paremeters instead of the parameters themselves, which reduces the operation errors and improves the positioning accuracy efficiently. During the estimation, the mass measurement data do not have to be saved, The data measured dynamically can be easily processed in real time. This simulation results indicate that it has positioning precision and high convergent rate.