INS/GNSS/Vision组合近距空中加油相对导航算法

INS/GNSS/Vision combined short-range air refueling relative navigation algorithm

针对空中加油近距对接阶段GNSS信号易受遮挡丢失的问题,考虑引入Vision导航系统,提出一种INS/GNSS/Vision组合的高精度容错相对导航系统,为空中加油近距对接任务提供高精度的双机相对导航信息。该系统采用联邦滤波器结构融合导航传感器信息,利用滑动窗口构建时变量测噪声的容错滤波结构,并基于子滤波器协方差的奇异值分块计算信息分配系数。仿真结果表明,所设计的容错导航系统与单独的INS/GNSS和INS/Vision系统相比,相对位置、速度和姿态的估计精度都有所提高;同时在GNSS出现软故障时,系统可以提供稳定、可靠的相对导航信息,有效提高了INS/GNSS/Vision组合近距空中加油相对导航系统的导航精度、鲁棒性和容错性,相对位置精度较基于常规故障隔离联邦滤波算法提高70%以上,满足了空中加油对接导航需求。

In view of the problem that the GNSS signal is easily blocked and lost during the short-range docking stage of air refueling,the Visual navigation system is considered to be introduced,and a high-precision,fault-tolerant relative navigation system of the combination of INS/GNSS/Visual is proposed to provide high-precision relative navigation information for the short-range docking task.The federated filter structure is used to fuse the navigation sensor information,and the sliding window is used to construct a fault-tolerant filter structure for time-variable noise measurement,and the information distribution factor is calculated based on the trace block of the sub-filter covariance.The simulation results show that the designed fault-tolerant navigation system has improved position,velocity,and attitude estimation accuracy compared with the separate INS/GNSS and INS/Visual systems.At the same time,when GNSS soft failure occurs,the system can provide stable and reliable relative navigation information,which effectively improves the navigation accuracy,robustness and fault tolerance of the INS/GNSS/Vision integrated relative navigation system for short-range docking stage of air refueling.The relative position accuracy is more than 70%higher than that of the federated filtering algorithm based on conventional fault isolation,which meets the requirements of air refueling docking navigation.