Research On The Application Of Adaptive Robust CKF Algorithm In Integrated Navigation

Integrated navigation technology can combine the advantages of different navigation systems and comprehensively use them to achieve better navigation results.Therefore,it has been widely used in navigation,intelligent transportation,mapping,weapon guidance,etc.Beidou navigation system is a new generation of global positioning system that will be developed by our country and will be networked.Although it can provide high-precision positioning information around the clock,it has the disadvantages that signals are easily interfered and the output frequency is low.As a commonly used navigation system,strapdown inertial navigation system has the characteristics of high output frequency but error accumulation,which is exactly complementary to BDS.When the system uncertainty increases and the observation is abnormal,The filtering accuracy of the standard CKF would decrease.To solve this problem,this paper does some research on these problems.The main work is divided into the fellowing aspects:(1)When the carrier is in a state of sudden change,the filtering accuracy of the standard CKF would decrease.To solve this problem,a strong tracking SVDCKF integrated navigation algorithm based on multiple fading factors(MST-SVDCKF)is proposed.The singular value decomposition(SVD)is introduced to replace the original Cholesky decomposition in standard CKF,which improves the numerical stability of the state covariance matrix decomposition iteration.Then,the multi-fade factor is employed to adjust the prediction state covariance matrix,which enables different filter channels to obtain different fade-out capabilities.In this way the strong tracking of the true state of carrier can be achieved.The simulation results show that the proposed algorithm has stronger adjustment ability and higher filtering precision than CKF and traditional STCKF.(2)Based on the strong tracking filter in the previous chapter,an adaptive robust SVDCKF integrated navigation algorithm(CTAR-SVDCKF)based on the chi-square test is proposed to solve the problem that the integrated navigation system has poor robustness when the statistical characteristics of process noise are uncertain and observations are abnormal.The algorithm first introduced the chi-square test to evaluate the system for the first time,and adjusted the prediction state covariance matrix based on the chi-square test value by using multiple fade factors;The chi-square test was introduced again to determine whether there is an observation anomaly,and affect the weighting matrix derived by Using M-estimation,which can then modify the innovation.The simulation results show that the proposed algorithm can effectively suppress the influence of process noise and observation anomaly on the system,and it is also applicable when the noise is normal and the observation value is sufficient.(3)A SINS/BDS information processing platform was designed.The data of sports car experiments were collected and the format of the collected data was converted first,and then the data was fused according to the established loose combination model and filtering algorithm.Finally,the comparison with the position information output by BDS proves that the information processing platform can effectively collect,process and fuse the experimental data.