一种基于多虚拟观测量的组合导航自主完好性监测方法

完好性用于导航系统出现故障的情况下提供及时的告警，是生命安全类用户需要考虑的重要性能指标。基于惯性导航系统（INS）辅助的卫星导航自主完好性监测算法，提出了一种利用INS构造3颗虚拟卫星观测量的组合导航系统自主完好性监测方法，通过构造视线方向相互垂直的3颗卫星，最大限度利用组合导航中INS的导航信息，在2颗可见星条件下就能实现故障检测。相比传统接收机自主完好性监测(RAIM)方法，该方法提高了检测算法可用性。在GPS单星座且INS定位误差方差σ_s=1m条件下，加入35m单星故障幅值，传统加权RAIM故障检测概率为48.51%，本文算法检测概率为95.21%，检测性能提升47%；与INS辅助的卡尔曼滤波残差检测法相比，在相同INS精度等级条件下该方法也具有更高的检测概率。此外，仿真分析了不同INS精度等级对该方法故障检测性能的影响，结果表明INS精度越高，该方法检测性能越好。

An autonomous integrity monitoring method of integrated navigation based on multiple virtual observations#br# #br#

Integrity can provide timely warning in the event of a navigation system failure, which is an important performance indicator for users to consider life safety. This paper is based on the autonomous integrity monitoring algorithm of satellite navigation assisted by the inertial navigation system (INS). An autonomous integrity monitoring method for integrated navigation system was proposed. This method uses INS to construct three-virtual-satellite observations, and can realize fault detection under the condition of two visible satellites by constructing three satellites perpendicular to each other in the line-of-sight direction and using the navigation information of INS to the maximum extent. This improves the availability of detection algorithm compared with the traditional receiver autonomous integrity monitoring (RAIM). The fault detection probability of traditional weighted RAIM is 48.51% under the condition of GPS single constellation and INS positioning error variance σ_s=1m with the addition of 35m fault of single satellite. The detection probability of the algorithm proposed in this paper is 95.21%, and the detection performance is improved by 47%. This method also has a higher detection probability than the INS-aided Kalman filtering innovation method with the same INS accuracy. In addition, the influence of INS precision on the fault detection performance of this method was simulated and analyzed. The results show that the higher the INS precision is, the better the detection performance of the proposed method is.