抗差自适应分步滤波算法在PPP/INS组合导航中的应用

针对常规抗差自适应滤波算法在PPP/INS组合导航应用中存在难以准确识别和分离观测粗差及运动异常对定位结果影响的问题,基于分类因子自适应滤波原理,提出了一种抗差自适应分步滤波算法。该算法首先执行第一步滤波,对状态模型异常信息进行隔离,仅对观测粗差进行诊断和抗差处理;然后在第一步滤波的基础上,执行第二步滤波,对状态模型异常进行诊断和自适应处理。算法分析表明,抗差自适应分步滤波算法可以准确地识别和分离观测粗差和运动异常扰动。实验结果表明,抗差自适应分步滤波算法能够进一步增强滤波算法抵抗观测粗差和运动异常扰动对滤波结果的影响,提高PPP/INS组合导航系统定位结果的稳定性和可靠性。

Two-step Robust Adaptive Filtering with Application to PPP/INS Integrated Navigation

Aiming at the problem that it is difficult to accurately identify and separate the effect of observation gross errors and kinematic model abnormality on the positioning results in PPP/INS integrated navigation applications by the conventional adaptive robust filtering algorithm, a two-step robust adaptive filtering algorithm is proposed based on the classification factor adaptive filtering principle. The algorithm includes two filtering stages: during the first stage of filtering, the abnormal information of the state model is isolated, and only the observation gross errors are diagnosed and robust. In the second stage of filtering, the state model abnormality is diagnosed and adaptively processed based on the filtering results of the first stage. The analysis results of the proposed two-step robust adaptive filtering algorithm show that the filtering algorithm can accurately identify and separate observation gross errors and kinematic model abnormality. The experimental results show that the two-step robust adaptive filtering algorithm can dramatically enhance the robustness of the filtering against the influence of observation gross errors and kinematic model abnormality and improve the stability and reliability of the PPP/INS integrated navigation system.