Research On Kalman Filter Algorithm Based On GPS/INS Integrated System

Integrated navigation technology has been greatly developed in our society today.How to improve the accuracy of the system is a problem that researchers all over the world consider.Some people start with hardware equipment and use higher-precision hardware systems to reduce errors,and some start from the algorithm level to reduce computational complexity and increasing algorithm stability.They all have their own advantages and disadvantages.They all have their own advantages and disadvantages.Based on the understanding of GPS system and INS system,this topic conducts research on GPS/INS tight integrated navigation system,and proposes a new and improved Unscented Kalman filter(UKF)algorithm:(1)In order to solve the problem of local singularity caused by the excessive number of sampling points in GPS/INS integrated navigation system,a Singular Value Decomposition-based Memory Unscented Kalman Filter algorithm(SVD-MUKF)is proposed,which is compared with the cubature Kalman filter(CKF)algorithm in positioning accuracy.The results show that the optimized SVD-MUKF filtering algorithm is more accurate and more stable than the CKF filtering algorithm.(2)Aiming at the problem of cumulative errors generated by the UKF algorithm after working for a long time,a memory attenuation factor is introduced into the UKF algorithm,and the filter attenuation factor is optimized in the case of multiple dimensions of the system,and the multiple memory attenuation filtering technology is used to keep the algorithm stable at all times.(3)In order to verify the effectiveness of SVD-MUKF filtering algorithm in practical application,a navigation experiment platform of SVD-MUKF filtering algorithm is built.The vehicle positioning and tracking experiment is carried out.The map environment includes complex terrain such as tall buildings,thick trees and roads,which meets the complex geographical conditions set by the experiment.The experimental conclusion proves that the SVD-MUKF filtering algorithm can effectively resist these disturbances and meet the expected requirements...