Study On Some Crucial Technologies Of BDS/SINS Integrated Navigation System

Beidou satellite navigation system(BDS)can provide highly reliable and highprecision navigation and positioning information,but the signal quality of BDS is easily influenced by occlusion or error factors.Strapdown Inertial Navigation System(SINS)is an autonomous navigation system.The signal quality can be free from interference,but the error accumulates over time,and initial alignment is needed before work.BDS and SINS have their own advantages and disadvantages,but their performances are complementary.The Beidou/SINS integrated navigation system,which is composed of organic combination,has better comprehensive performance than its own single navigation,which can effectively improve the overall performance of navigation.Based on the related problems of BDS and SINS,some key technologies of Beidou /SINS system are studied.An improved cubature Kalman filter algorithm is studied for the initial alignment problem of SINS.Firstly,the model of SINS initial alignment is established and analyzed.Then,according to the characteristics of the initial alignment model equation,the cubature kalman filter is used in the time update phase of the algorithm,and the SVD decomposition is introduced to assist the covariance decomposition to ensure numerical stability and the L-M algorithm is introduced to ensure convergence.While kalman filter is used in the measurement update phase of the algorithm.Finally,the experimental results show that the improved cubature Kalman filter algorithm improves the alignment accuracy and optimizes the computational efficiency.Aiming at the low-cost SINS attitude calculation process,the noise interference is large,the precision is not high,and the drift is difficult to control.An improved Kalman filter algorithm based on PID control is adopted.Firstly,the attitude data of the accelerometer and the gyroscope are respectively solved.Then,based on the kalman filter,the Joseph-type covariance is introduced to avoid the singularity and the innovation is introduced to adjust the observed noise covariance,that is,the improved kalman filter is formed.Then,the modified kalman filter is used to optimize the fusion output of the PID-adjusted accelerometer and gyroscope attitude information.Finally,a comparative experiment is carried out.The results show that the improved Kalman filter algorithm based on PID control improves the accuracy of attitude calculation.An improved Kalman filtering algorithm based on stationary wavelet transform is adopted to solve the problem of low accuracy and large fluctuation of system state estimation caused by noise of observation data in Beidou/SINS integrated navigation system.Firstly,the algorithm denoises Beidou data by stationary wavelet transform(SWT).Then,based on Kalman filter framework,H? filter is introduced to improve the stability,that is to say,an improved algorithm is formed.Then,the improved algorithm is used to fuse the SINS data and the denoised Beidou data and estimate the system state.Finally,a comparative experiment is carried out.The results show that the improved Kalman filter based on stationary wavelet transform has better performance.