Research On Initial Alignment Technology Of MEMS SINS

With the continuous improvement of inertial devices and computer performance,strapdown inertial navigation system has become the mainstream in the field of navigation.Strapdown inertial navigation system with high precision devices has been widely used in various fields.With the advantages of low cost and small size,the strapdown inertial navigation system based on the micro-electro-mechanical system(MEMS)has attracted much attention in the development of the inertial navigation system.Therefore,the strapdown inertial navigation system based on the micro-electro-mechanical system(MEMS)has emerged and developed rapidly.However,due to the low precision of the MEMS inertial devices,the performance of the navigation system is easily affected,especially for the initial alignment,which limits the development of the MEMS strapdown inertial navigation system.Based on this,the initial alignment technology of MEMS strapdown inertial navigation system is deeply studied in this paper.First,this paper studies the method of alignment aided by magnetometer on stationary base: The initial heading angle and the observed quantity of Kalman filter calculated by magnetometer are used to assist alignment.The simulation results show that the scheme meets the requirements of alignment accuracy,but it also has some limitations.In order to improve the value of the initial alignment scheme in practical application,this paper proposes an initial alignment based on the application of biaxial rotation technology.Then,the constant error can be modulated to a periodic change signal with zero mean value,thus some of the device errors could be eliminated and the initial alignment accuracy could be improved.In the scheme aided by magnetometer,three methods for calculating strapdown matrix are presented,including one direct calculation of heading angle by magnetometer and two different kinds of indirect calculations by constructing vector matrices.Kalman filter is used for precise alignment,and heading angle error is used as measurement.The simulation and experimental verification of real data are carried out.The experimental results show that the initial alignment scheme assisted by magnetometer has about 5 minutes horizontal attitude angle errors and the heading angle error is less than 1 degree.Finally,in order to test the anti-jamming capability of the magnetometer,the heading angle error drift is calculated by the magnetometer data affected by the iron block,which verifies that the output of the magnetometer is susceptible to environmental impact.It paves the way for the design of an autonomous alignment scheme based on biaxial rotation.In view of the low precision of MEMS inertial devices,this paper firstly designs the dual-axial rotation scheme based on three factors: rotation order,rotation speed and turn-around time.Then error analysis on the determined dual-axial rotation to verify the effect of rotation modulation is made.After that,a set of dual-axial rotation scheme is designed: in the coarse alignment,the compass alignment method is simplified by omitting the realignment process,and the attitude angle of the carrier is estimated directly;after that,the observability of the system is analyzed,the optimal state variables are selected,and Kalman filtering and an improved Saga-Husa adaptive filtering are used for fine alignment.The simulation results show that both filtering methods achieve the ideal alignment accuracy.At the end of this paper,in order to test the performances of the initial alignment schemes under actual working conditions,the semi-physical simulation experiment assisted by magnetometer and the rotating table experiment with two axes are designed.At last,the practicability of these two initial alignment schemes is proved by the simulation of experimental data.