Research On Multi-source Navigation Information Fusion And Fault Detection Algorithm

Thanks to the continuous development of modern science and technology,the types of navigation systems that can be used for integration are increasing,and the requirements of the carrier for navigation systems are also increasing.The combination of information sources from different types of navigation systems can overcome their inherent defects and realize the complementary advantages of each subsystem.The multi-source information fusion algorithm can fuse the information of the subsystem and ensure the navigation accuracy of the system,which is the working basis of the navigation system.Fault detection algorithm can detect fault information in time,which is an important guarantee for system reliability.In order to obtain the all-weather and all-terrain integrated navigation system,make the navigation system obtain more accurate positioning information,and improve the robustness and reliability of the system,this paper mainly studies the multi-source navigation information fusion algorithm and fault detection algorithm.Firstly,the importance of multi-source navigation system integration is analyzed,and then the basic principle of multi-source navigation system is introduced.Through the derivation of the basic knowledge of inertial navigation system and multi-source information fusion theory,the mathematical modeling of multi-source navigation system is completed.Secondly,the time synchronization problem of multi-source navigation system is studied,and the information delay caused by communication,operation and other reasons in the actual work of the system is emphatically studied.In view of the problems existing in the existing time synchronization methods,the delay state is added as an extension to the state equation and a new mathematical model is constructed,so as to achieve more accurate delay estimation,and the method is verified by simulation experiments.Thirdly,the robustness of multi-source navigation system is studied.In order to solve the problem that Kalman filter(KF)algorithm relies too much on the accuracy of noise matrix,the unbiased finite impulse response filter(UFIR)is introduced to improve the robustness of the sub-filter of the federal filter.On the one hand,aiming at the problems existing in the existing optimal window length estimation algorithm,an online optimal window length estimation algorithm is proposed to reduce the estimation time and improve the operation efficiency of the algorithm;On the other hand,in order to solve the problem of low filtering accuracy of UFIR algorithm,the adaptive interactive multiple model(IMM)algorithm framework is used to weight the estimation results of UFIR and KF algorithm,which improves the robustness of the algorithm and ensures the filtering accuracy of the algorithm.The effectiveness of the algorithm is verified by simulation and analysis.Finally,the fault detection algorithm of multi-source navigation system is studied,focusing on the slow-varying fault characteristics,and an adaptive fault detection algorithm based on generalized relative entropy theory is proposed.In view of the low accuracy of slow-varying fault detection,the difference between the prior Gaussian probability distribution of the dynamic model of each subsystem and the prior Gaussian distribution obtained from the update step of the federated sub-filter is used as a parameter by using the generalized relative entropy theory.The weight distribution is carried out by measuring noise characteristics and system noise characteristics,and a fault detection method suitable for multi-source navigation system is designed.At the same time,in view of the problem that the traditional federal filtering algorithm under fault conditions will introduce faults,an adaptive allocation factor algorithm that can adapt to the characteristics of subsystems under fault conditions is designed,and the effectiveness of this method is proved by simulation.