GPS单点定位结果的动力平滑精度受到单点定位误差及动力学模型误差的影响,本文将抗差LS—M滤波算法应用于动力平滑中以控制上述两项误差的影响,着重讨论了LS—M滤波在动力平滑单点定位结果中的作用,并与自适应滤波进行比较。实验结果表明,LS—M滤波在状态噪声补偿较差的情况下可以提高滤波精度,在一定程度上控制了历史观测异常对定轨结果的影响。
The accuracy of dynamic smoothing orbit determination based on GPS single point positioning is decreased due to both single positioning error and dynamic model error. This paper mainly discusses the function of LS - M filtering in dynamic or- bit smoothing, and compares it with adaptive filtering. The result shows that LS - M filtering can improve the filtering accuracy when state noise compensation is not good enough, and can control the influences of observation outliers to a certain extent.