面向城市复杂环境的GNSS/INS高精度图优化算法

The high-precision factor graph optimization algorithm of GNSS/INS for urban complex environment

在城市复杂环境下,GNSS接收机易受到建筑物遮挡、多路径效应等多种因素影响,导致信号出现粗差或者拒止情况,从而对GNSS/INS组合导航系统的精度和鲁棒性造成影响。提出了一种具备粗差在线检测的GNSS/INS图优化组合导航算法,提高城市环境条件下的组合导航系统性能。基于信息之间存在关联性的特点,设计了一种卫星信号滑窗粗差检测与拟合替换算法,抑制卫星粗差影响;构建了GNSS位置、速度因子和改进的IMU预积分因子,实现了组合导航信息非线性优化。仿真和车载数据试验表明,针对卫星信号中存在粗差的情况,所提算法的定位精度相比扩展卡尔曼滤波和传统图优化算法提升90%以上,可以辅助导航系统获得较好的状态估计效果;针对GNSS拒止的情况,该算法的位置定位精度相比扩展卡尔曼滤波算法提升30%以上。

In the complex urban environment,GNSS receivers is easily affected by various factors such as building occlusion and multipath effects,lead to gross errors or rejection of the signal,which easily affects the accuracy and robustness of the GNSS/INS integrated navigation system.A factor graph optimization algorithm of GNSS/INS with gross error online detection is proposed to improve the performance of the integrated navigation system under the interference condition of urban environment.Based on the characteristics of correlation between information,a sliding window gross error detection and fitting replacement algorithm for satellite signals is proposed to suppress the influence of satellite gross errors.The GNSS position,velocity factor and improved IMU pre-integration factor are constructed to realize nonlinear optimization of integrated navigation information.The simulation and vehicle experiments show that the positioning error of the proposed algorithm is reduced by more than 90%compared with the extended Kalman filter and the traditional factor graph optimization algorithm in the case of gross errors in satellite signals,which can assist the navigation system to obtain a better state estimation effect.In the case of GNSS rejection,the positioning error of the algorithm is reduced by more than 30%compared with the extended Kalman filter algorithm.