Fault Detection And Reconstruction In Integrated Navigation System

As the weapons and other aircraft systems' requirements on the positioning accuracy and reliability increasing.Integrated navigation system concluding many navigation systems becomes the most ideal system.The fault detection algorithm of the INS/GNSS/CNS is studied in this paper in order to further improve the accuracy and reliability.Based on the common method of residual chi-square test,this paper proposes an improved sequential residual probabilistic ratio test method,which can detect the fault of the integrated navigation system simply and accurately.This method does not need to consider the length of the data window on common soft-fault detection of the integrated navigation system.The simulation results of INS/GNSS/CNS integrated system show that the proposed algorithm is more sensitive to soft faults and can improve the soft fault detection capability of integrated navigation system.Since the INS/GNSS/CNS integrated navigation system position feedback is mainly from the satellite navigation system,the integrated navigation system positioning accuracy will be decreased because of the INS/GNSS's isolation when INS/GNSS subsystem is detected failed,therefore,we use the improved kernel neural network with the genetic algorithm to train the model when the system is normal,and use the trained model to locate the solution in case of system failure.The simulation results of the integrated navigation system show that the proposed method can maintain the high precision of navigation and improve the fault tolerance of the integrated navigation system when the INS/GNSS satellite navigation subsystem is faulty.