Research On The Technique Of Single-axis Rotating FOG Strapdown Inertial Navigation System

In strapdown inertial navigation system, the continuous research of the gyroscope and the accelerometer promote the rapid development of inertial measurement unit (IMU).However, the higher the precision of the IMU, the greater the cost of further improving the precision of IMU. Due to the inertia sensitive component of a high precision manufacturing is very difficult and expensive. Therefore, applying advanced system technique to improve the navigation precision without developing a new generation of Inertial sensing element has always been a vital task in the field of INS. In this paper, viewed the single shaft rotary fiber strapdown inertial navigation as the research object, to study the error suppression technique.The main work of this paper includes the following aspects.In this paper, we first study the self calibration technique of single axis rotating optical fiber SINS, and establish the Kalman filter model with the error equation of the system and the error model of the IMU. According to the error equation, the observability of the system error parameters is analyzed. In order to more intuitive to rotation scheme is able to improve each state variable of the degree of observation were evaluated using based on SVD theory observability analysis method respectively in static state and IMU around the rotation axis of rotation of the system observability analysis, based on the observability analysis results, this paper puts forward four eight position order calibration path, and the self calibration scheme is simulated.Secondly, the inversion scheme of single axis rotating optical fiber SINS is studied.Based on the classic four position rotation, the angular velocity, angular acceleration and stop time are analyzed. Laplasse transform is used to solve the error equation, the oscillation error of the system is analyzed, and the effect of rotation speed on the single axis rotation SINS is studied. Modeling the angular velocity change of acceleration and deceleration process, the error modulation of the whole velocity variation process is analyzed, and the influence of angular acceleration on the system navigation accuracy is analyzed. Concludes with a discussion on the influence of the dwell time of rotation modulation effect, due to the stopping time can’t be determined by simple mathematical model to determine, so by simulation analysis to compare the different effect of cycle of location error.At the end, Studied single shaft axial rotation sins gyro drift precision identification technology, axial drift error propagation analysis and the high precision modeling, and proposes a kind of machine learning uniaxial sins axial gyro drift identification method based on. By constructing a large number of feature representation gyro drift and latitude error. By means of principal component analysis feature compression and filtering to improve the operating speed and the effect is described. Through the simulation verification of the constructed model has excellent generalization approximation ability, can better identification axial drift and compensation, under the simulation conditions, the compensation accuracy better than 0.0005 °/h.