Research On Compensation And Calibration Technology Of MEMS Magnetometer

MEMS magnetometer as an important sensor for geomagnetic navigation,can measure the strength of the geomagnetic field,and can also solve the important heading angle information.In actual using,the magnetometer is extremely susceptible to environmental interference,and there are inevitable self-errors in the manufacturing process.Therefore,the magnetometer must be corrected before using,and the accuracy of the correction directly determines the accuracy of heading angle.How to use an effective algorithm to achieve accurate correction of the magnetometer to improve the system's magnetic field measurement accuracy and heading angle output accuracy is the main content of this paper.This paper focuses on the analysis and modeling of magnetometer error,off-line correction algorithms and online correction algorithms.The main research contents are as follows.Firstly,the basic principle of the magnetometer and heading angle determination are introduced.The self error,environmental error and installation error of the magnetometer are analyzed and mathematical models are established.The total error model and compensation model of the magnetometer are derived.The magnetometer correction algorithms based on ellipsoid fitting,six-position rotation and attitude independence are introduced,and their advantages and disadvantages are analyzed by simulation experiments.Secondly,the traditional particle swarm optimization algorithm for magnetometer correction is introduced.Aiming at the problems of less estimation parameters,a particle swarm optimization algorithm for correcting magnetometer based on MEMS gyroscope is proposed The number of estimated parameters can be increased from 9 to 12,and the full parameter estimation of the error matrix is realized.The random drift particle swarm optimization algorithm is introduced to improve the dynamic adaptability of the algorithm.The superiority of the algorithm is verified by simulation and actual experiments.Then,considering the problem that the accuracy of traditional algorithm's estimation parameters is low,a particle swarm optimization magnetometer correction algorithm based on the double objective function is proposed.The von Neumann topology is used to improve the global search ability of the algorithm.The results of the simulation and practical test show that the algorithm improves the parameter estimation accuracy rate to over 95%and has higher robustnessFinally,an online correction algorithm of magnetometer based on INS/GPS assisted is proposed for the changeable magnetic field environment.The online calibration of the magnetometer is realized by using the extended Kalman filter algorithm under the GPS information.After the GPS is unlocked,the magnetometer after the online compensation is used to provide the heading angle information to maintain the orientation of the system.Considering the influence of three motion trajectories of straight line,rectangle and "8" on the correction results,the simulation and the sports car experiments show the advantages and disadvantages of each,and verify the necessity of the second correction.The Qt-based magnetometer correction software is designed to realize the functions of data import and export,simulation data generation and algorithm operation,which improves the engineering practicability of the algorithms.