Design On Fault Detection And Reconstruction Algorithm Of INS/GNSS/CNS Integrated Navigation System

High altitude long-time solar unmanned aerial vehicle(UAV)has a long flight time and complex working environment,that means its navigation system should be highly reliable and high-precision.Also,this system should have redundancy and fault tolerance ability.Thus,the single navigation system is difficult to meet the needs of UAV long duration flight.It is necessary to develop a multi-sensor integrated navigation system;and it is required to be able to detect the fault of a subsystem quickly to ensure the reliability of the navigation system.Therefore,a SINS/CNS/GNSS integrated navigation system is designed in this paper,so do the fault detection and system reconstruction algorithm of the integrated navigation system.The main contents of this paper are as follows:(1)The conventional SINS/CNS observation model ignores that the installation angle measurement error between the CNS sensor and the SINS sensor,which makes the decrease of the SINS/CNS subsystem's precision.a improved SINS/CNS model is designed,in which the influence of installation error angle on the system is considered.Compared to the conventional model,this model extends its state dimensions to 18 from the conventional 15 state dimensions model.The EKF algorithm is used for simulation and comparison experiments for those two models,and the results show that the precision of the improved model is higher than that of the conventional model(2)A SINS/CNS observation model based on star light vector is designed to solve the problem that the conventional SINS/CNS model needs the CNS's sensor can be badly affected by the sunlight.The original measurement value of star sensor is used to update the measurement value,and the EKF algorithm is used for simulation,the results show that the INS/CNS can still work normally when the number of observations is less than 3 with this model.(3)In view of the sensor failure in the integrated navigation system,an improved SPRT fault detection algorithm is designed.By setting zero setting operation and window,the problem of fault detection delay of traditional SPRT algorithm is avoided.The simulation results show that the algorithm has a good detection effect on slow and sudden change faults.(4)An adaptive filtering algorithm for weight factor assignment based on fault probability information is designed.The fault probability of each subsystem is obtained by fuzzing the threshold value of the fault detection algorithm,and the weight factor of each subsystem is changed in real time according to the fault probability.The optimal estimation of the integrated navigation system can be gotten through those weight factors of subsystems.Simulation results show that the algorithm can effectively reduce the impact of faults on the system and improve the reliability of integrated navigation system...