基于矢量化检测联邦滤波的INS/BDS/地磁组合导航容错方法

组合导航能够将多种类型的导航信息进行结合,实现优势互补,因此成为了目前导航应用领域的主要发展方向。然而,导航信息的增多势必会引入更多的风险源,从而降低导航系统的可靠性。基于联邦滤波的容错方法是目前抑制故障信息影响的主要解决手段,但是现有的故障容错方法普遍采用统一的检测机制,没有根据各个导航子系统的误差传播特性针对性地构建故障检测模型,因此会引起较高的误警率与漏检率。针对上述问题,提出了基于矢量化检测联邦滤波的INS/BDS/地磁组合导航容错方法。通过构建面向INS/BDS/地磁不同导航信息的故障检测函数,能够实现更加准确的矢量化信息分配,从而可以有效避免可用导航信息的损失以及故障导航信息对整体系统的影响。仿真结果表明,提出的方法可以有效隔离不同类型的故障信息,并减小其对无故障导航信息及整体系统的影响,从而提高了组合导航的精度和可靠性。

Fault-tolerant approach for INS/BDS/geomagnetic integrated navigation based on vector inspection federal filtering

Integrated navigation, which combines multiple types of navigation information to achieve complementary advantages, has become the main direction of development in the field of navigation applications. However, the increase of navigation information is bound to introduce more sources of risk, reducing the reliability of the navigation system. The fault-tolerant method based on federated filtering is currently the main solution to suppress the impact of fault information. Nevertheless, the existing fault-tolerant methods generally adopt a unified detection mechanism, without building fault detection models according to the error propagation characteristics of each navigation subsystem, which may cause higher false alarm rates and missed detection rates. To address these problems, this paper proposes an INS/BDS/genmagnetic integrated navigation fault-tolerant method based on vectorized detection and federated filtering. By constructing fault detection functions for different navigation information of INS/BDS/genmagnetic, more accurate vectorized information allocation can be achieved, thereby effectively avoiding the loss of available navigation information and the impact of fault navigation information on the overall system. Simulation results show that the proposed method can effectively isolate different types of fault information, reduce their impact on fault-free navigation information and the overall system, and improve the accuracy and reliability of integrated navigation.