Research On Error Suppression Method Of MEMS INS/BDS Vehicle Integrated Navigation

With the development of Micro-Electro-Mechanical System(MEMS)technology,MEMS inertial sensors represented by gyroscopes and accelerometers have the characteristics of low cost,small size,low power consumption,and high reliability.Therefore,the inertial navigation system(INS)composed of MEMS inertial sensors has been widely used in the field of civilian consumption and large-scale professional equipment,which greatly expands the application scenarios of inertial technology.Due to the complementary advantages of MEMS INS and satellite navigation(such as Beidou satellite navigation system,BDS),the combination of them has become a common integrated navigation method.In order to obtain high-precision navigation and positioning information,this paper takes the vehicle integrated navigation as the research object,analyzes and studies the error of the inertial sensor and various robust algorithms of the integrated navigation,and designs the calibration method of the inertial unit and robust algorithms of integrated navigation.The correctness and effectiveness of the algorithm in this paper are verified by experiments.The main research contents of the paper are as follows:Firstly,aiming at the problem of deterministic error calibration and compensation of MEMS accelerometers,this paper proposes a MEMS triaxial acceleration calibration method based on an improved adaptive genetic algorithm.In order to fully consider the error incentive,design a 24-position transposition scheme to collect calibration data.Using six-position rough calibration method to determine the approximate search interval of the genetic algorithm,and the genetic operator that changes dynamically with the evolution times to find the optimal solution iteratively.This method can improve the problems of low calibration accuracy or incomplete calibration parameters caused by the traditional iterative method,which is easy to fall into local convergence.Secondly,to address the problem that the deterministic errors of MEMS gyroscopes are easily coupled with accelerometer errors,and to achieve higher accuracy calibration compensation,this paper proposes a two-step correction method based on the MEMS three-axis gyroscope calibration method.This method which firstly adopts the six-position correction method to calibrate the 12 error parameters of the accelerometer,the scale factor and static zero offset of the gyroscope.Then,assisting the non-orthogonal errors of the gyroscope for system-level calibration.The method takes full account of the gyroscope error parameters,reduces the system dimensionality of the Kalman filter and enables fast and accurate identification of each error without precision equipment.Then,for the problems of model dynamic interference and measurement outliers in the integrated navigation,this paper research the commonly adaptive filtering and Bayesian filtering methods,designs and improves the filtering algorithm of the integrated navigation system.This paper proposes a multifactor adaptively robust filtering method.This method makes full use of the position dilution of precision of satellite signals as the first-level detection,further constructs a robust adaptive factor and equivalent weight matrix.It improves the positioning accuracy of vehicle integrated navigation in complex environments.Finally,based on the above method,the calibration method was tested on inertial sensors of different grades.The experimental results show that the RMSE of the three-axis accelerometer reduced to 1.33mg after calibration.Compared with the classical Newton method,the RMSE is improved by 22.2%.After the calibration of the three-axis gyroscope,the pitch angle error reduce to 0.624°,and the roll angle error reduce to 0.67°.The vehicle experiment of the integrated navigation system based on MEMS INS/BDS platform shows that the positioning accuracy RMSE of the multi-factor adaptive robust filtering is improved by 43.58%,26.39%and 71.36%in three directions respectively which compared with the standard Kalman filter.The above results show that the various error suppression methods for inertial sensors and integrated navigation studied in this paper can effectively improve the positioning accuracy of MEMS INS/BDS integrated navigation,and provide method support for high accuracy and low-cost vehicle integrated navigation and positioning.