针对光纤陀螺捷联惯导(FOG SINS)/GPS组合导航系统实际工作环境中,由于系统噪声与量测噪声模型发生变化而带来的滤波器发散的问题,提出一种新型模糊自适应Kalman滤波器(FSHAKF)。通过引入IMU精度因子与GPS水平精度因子,构造模糊推理系统(FIS),实时更新自适应参数,有效地解决了传统Sage-Husa自适应滤波器(SHAKF)估计模型不准确、系统噪声与量测噪声无法同时估计以及滤波器长时间易发散的问题。仿真实验表明,本文提出的FSHAKF算法相较于SHAKF算法,估计精度得到明显提高,且避免了滤波器的发散。 Aiming at the problem of divergence of filter due to the change of system noise and measurement noise model in the actual working environment of FOG SINS / GPS integrated navigation system, a new fuzzy adaptive Kalman filter (FSHAKF). By introducing IMU precision factor and GPS level accuracy factor, the fuzzy reasoning system (FIS) is constructed and the adaptive parameters are updated in real time, which effectively solves the problems of inaccurate estimation of the traditional Sage-Husa adaptive filter (SHAKF) estimation model, system noise and measurement Noise can not be estimated at the same time and the filter is easy to diverge for a long time. The simulation results show that compared with the SHAKF algorithm, the proposed FSHAKF algorithm can significantly improve the estimation accuracy and avoid the divergence of the filter.