Fast Self - Calibration of Fiber - Optic Strapdown Inertial Navigation System
Dingjie Xu,Pan Jiang, Ya Zhang, Shiwei Fan,Guochen Wang, Harbin Institute of Technology, China
Location: Big Sur
Gyro bias is the main error source of the strapdown inertial navigation system (SINS), and the bias of the accelerometer determines the horizontal error of the SINS. So we propose a fast self-calibration method, which can accurately calibrate the accelerometer and the gyro bias within 15 minutes .The observability of the SINS is effectively improved by changing the position of the inertial measurement unit (IMU) in Kalman filter. Through the 160 experiments, it can be found that the calibration of the accelerometer and the gyro bias has reached the requirement of accuracy while satisfying the fastness.
It is found that changing the position of the inertial measurement element can change the system matrix in the error model of strapdown inertial navigation system, which can improve the observability of the system and make the system completely observable. Thus completing the horizontal direction accelerometer and gyro zero deviation calibration.
Its essence is to change the heading angle of 180 degrees, so that the device error in the geographical system on the projection matrix changes, thus changing the system error model of the system matrix, improve the system's observability. The theoretical analysis shows that the change of the projection matrix of the device error on the geographic system when the IMU system is rotated 180 degrees is equivalent to the 180 degree change of the carrier heading, that is, the IMU system can rotate 180 degrees to achieve the best effect.
So we designed the fast self-calibration scheme for
(1) In the first position for 3min, and then around the z axis rotate 180 degrees within 30s.
(2) In the second position still 690s;
(3) Using the fast self-calibration algorithm to deal with off-line data, get gyro and accelerometer horizontal direction zero bias.
In this paper, the rapid self-calibration algorithm for the relevant simulation analysis to verify the effectiveness of the algorithm and the previous analysis of the conclusions. The simulation experiment mainly includes the following four cases: no error source, the existence of the error of the translocation mechanism, the existence of device noise, the error of the translocation mechanism and the device noise are both present. In addition, according to the above calibration process, we collected 160 sets of actual data for off-line processing, and the calibration results and schism calibration results were compared. The stability of the zero-deviation calibration results is consistent with the characteristics of the device, which meets the requirements of the inertial device. In the 160 self-calibration experiments, the zero bias of the gyro and the accelerometer is calibrated within 15min to meet the fast requirement. The correctness of the proposed fast self-calibration method is verified.
In this paper, the position of the inertial measurement element is changed in Kalman filter, which improves the observability of the strapdown inertial navigation system effectively. The zero drift of the accelerometer and the gyro can be accurately calibrated within 15 minutes, and the effects of the random drift of the gyro and the accelerometer and the turntable error on the calibration result are analyzed. Through the simulation and experimental verification, the zero deviation of the accelerometer and the gyroscope meet the requirement of accuracy while satisfying the fastness, and verified the feasibility of the scheme.