基于速度约束与模糊自适应滤波的车载组合导航

针对车载组合导航系统中卫星信号易受遮挡而引起导航精度降低问题，提出采用车辆速度约束条件辅助的组合导航方案。利用车辆正常行驶过程中侧向和天向速度为零作为虚拟观测信息，推导得到卫星信号失效时组合导航滤波量测方程；考虑到Kalman滤波过程中量测噪声协方差矩阵难以获取，推导给出一种新的自适应Kalman滤波（ADKF）算法，该算法计算新息序列实际协方差与理论协方差比值后，利用模糊推理系统（FIS）自适应调节量测噪声协方差矩阵大小；通过光纤捷联惯性导航系统（SINS）进行了验证试验。结果表明：卫星信号失效时，虚拟速度组合能够提高SINS定位精度，其纬度最大误差由41.33 m减小为8.61 m，且采用FIS-ADKF组合导航算法时3个方向 位置精度相比标准Kalman滤波算法提高了60%以上，验证了所提出算法的有效性。

Vehicle Integrated Navigation Based on Velocity Constraint and Fuzzy Adaptive Filtering

The satellite signal is easily obstructed to reduce the positioning precision of the integrated navigation system for the vehicle integrated navigation system. An integrated navigation scheme is proposed, which is assisted by the vehicle velocity constraints. The Kalman filter measurement equation is deduced by taking the side and up velocities of normal running vehicle be zero. A new adaptive Kalman filtering (ADKF) algorithm is deduced in consideration that it is difficult to determine the statistical characteristics of integrated navigation measurement noise. After calculating the ratio of actual covariance to theoretical covariance of innovation sequence, the fuzzy inference system (FIS) is used to adjust the Kalman's measurement noise covariance matrix adaptively. The verification experiments were carried out by using fiber optic strapdown inertial navigation system (SINS). The experimental results show that, when the satellite signal is invalid, the integration of virtual velocities can improve the positioning accuracy of SINS, and the maximum latitude error is reduced from 41.33 m to 8.61 m. The positioning accuracies of the vehicle's three directions calculated by the proposed FIS-ADKF integrated navigation algorithm are more than 60 percent higher than those calculated by the standard Kalman filtering algorithm, which verifies the effectiveness of the proposed algorithm. Key