Research On Systematic Calibration Of Strapdown Inertial Navigation System Of Fiber-Optic Gyroscope

This paper mainly researches systematic calibration as well as discreet calibration offiber optical gyro（FOG） stripdown inertial navigation system(SINS).The principle of SINS is to calculate the location and attitude of the carrier fromcertain initial data and the real-time carrier’s motion information provided by inertialmeasuring units(IMU)(gyroscopes and accelerators) which are directly installed on the carrierbody. However, as an autonomous system, the SINS, in the process of navigation, suffersfrom multiple system errors which are not able to be corrected by external information andthus impair the accuracy of the system outputs. The errors induced by IMU are the principalerrors. Therefore it is the key to increase the accuracy of the IMU in order to increase theoverall system accuracy. This paper mainly researches the methods to determine the errorterms of the IMU by calibration technology, as well as the algorithm to compensate the errorterms in SINS so as to increase the navigation accuracy.Firstly, the principle of discreet calibration method is discussed. Discreet calibrationis the most widely used calibration method because of its simple principle and flexibility inmodeling. However, due to some inevitable flaws, the discreet calibration method suffersfrom certain errors which badly affect the navigation outputs. The analysis of the effects ofdiscreet calibration method on the SINS is presented in this paper and a simulation is run tovalidate the analysis.Because of the multiple fallacies of discreet calibration, a new and innovatedcalibration method--the systematic calibration, whose theoretic core is Kalman filter, ispresented and profoundly researched in this paper. The principle of systematic calibration isas follows. The system measuring errors of SINS can be obtained by comparing the realvalues from the turntable and the navigation system outputs that are calculated from thegyroscopes and accelerators outputs. Regarding the system measuring errors as observervariables and the to-be-calibrated error terms of the IMU as state variables, the systematiccalibration algorithm utilizes the Kalman filter to achieve the unbiased estimation of the errorterms of gyroscopes and accelerators. In the light of this calibration method, all error terms ofthe IMU can be estimated with high accuracy, and therefore the accuracy of the SINS can beimproved by compensating those errors. Then, through considering these measuring errors as observer variables and sensor errorsas state variables in Kalman filter, an unbiased estimation is proceeded on the error terms ofinertial sensors. Thus all error terms of inertial sensors are accurately estimated and hence theerror compensation for SINS is able to be performed. Meanwhile, observability of the systemis analyzed and the calibration path is designed before the entire systematic calibrationalgorithm is simulated in MATLAB. The simulation result certifies the effectiveness of thismodel and feasibility of the algorithm which gives estimation of error terms.Finally, an experiment on turntable is carried out to compare systematic calibration anddiscreet calibration. And then the experiment results are applied to navigation calculationtogether with other actual data. The experiment results clearly demonstrate that systematiccalibration outperforms discreet calibration both on accuracy and on speed of convergence.Therefore, systematic calibration deserves more research efforts in the future.