抗差自适应EKF在INS/GNSS紧组合中的应用

针对惯性/卫星紧组合导航中卫星观测数据存在粗差，影响组合导航系统定位精度，同时考虑到卫星少于4颗和卫星几何分布不佳对预测残差构造自适应因子的影响，提出了一种抗差自适应EKF紧组合算法。该算法给出惯性/卫星紧组合状态方程与观测方程，抗差等价权因子和预测残差法构造自适应因子的计算公式，并给出卫星少于4颗和卫星几何状态分布不佳情况下自适应因子的计算方法。通过车载实测数据对算法进行验证与分析，实验结果表明，基于抗差自适应EKF的惯性/卫星紧组合算法可有效削弱卫星粗差观测值的影响，在可见卫星数少于4颗和卫星几何分布不佳的状态下，依据该算法获取的系统导航精度得到进一步提高。

Application of Robust Adaptive EKF in INS/GNSS Tight Combination

In inertial navigation system/global navigation satellite system (INS/GNSS) integrated navigation system, the positioning accuracy of integrated navigation solution will be degraded by the outliers of GNSS measurements. The predicted residuals and the setup of adaptive factors are affected if less than four satellites are observed and the satellite geometric distribution is poor. Therefore, a tightly coupled integration algorithm based on robust adaptive extend Kalman filter (EKF) is proposed in this paper. The state transition equation and observation equation of the tightly coupled INS/GNSS system are introduced. The equations of robust equivalent weight factor and adaptive factor constructed by using prediction residual method are derived. In addition, the adaptive factor calculation under the condition of less than four satellites and poor satellite geometric distribution is conducted. The proposed algorithm is verified and analyzed by using land vehicle test data, the experimental results show that the robust adaptive EKF INS/GNSS tightly integration method can effectively reduce the influence caused by satellite outliers. With the proposed algorithm, the navigation accuracy can be further improved under conditions of less than four satellites observable and poor satellite geometric distribution.