Research On Calibration Of Laser Strapdown Inertial Navigation System In Moving Base

The output of the inertial device inevitably contains various errors.The task of calibration is to eliminate these errors as much as possible,thereby improving the navigation accuracy of the inertial navigation system.At present,more new requirements have been put forward for calibration technology,and decreasing the calibration time and realizing the disassembly-free online calibration under the conditions of moving bases have become the research focus of calibration technology.The ship-borne SINS will inevitably be affected by wind and waves,which will be interfered by angular motion and line motion.Therefore,the difficulty of online calibration is greatly increased.The paper relies on a laser gyro SINS(LSINS)with a biaxial rotation mechanism to accomplish the following research work:(1)Online calibration method in mooring state is studied and verified by experiment: Ship motion in mooring state is modeled,and the track generator is designed.A high pass IIR filter is designed to obtain the velocity error,which is used as the observation of the Kalman filter.,so that the full process of calibration does not require external devices to provide reference information.The calibration results of “velocity matching” online calibration method in mooring state,“position matching” online calibration method in mooring state and “velocity + position matching” online calibration method in mooring state are compared through a series of experiments.The results indicate that this online calibration method in mooring state which uses “velocity matching” is the best and it can estimate all 24 error parameters including the bias,scale factor error and alignment error of the gyro and accelerometer accurately.The precision of the method can also reach the precision of the systematic calibration method in stationary base.(2)Rapid online calibration method in constant-speed moving state is studied and verified by experiment: Ship motion in constant-speed moving state is modeled,and the track generator is designed.The velocity error and the position error,which are used as the observation of the Kalman filter,is obtained by external devices.Observability analysis is performed,and it is theoretically proved that only the system which uses “velocity + position matching” is completely observable.The results of a series of experiments indicate that rapid online calibration method in constant-speed moving state which uses “velocity + position matching” can rapidly and accurately estimate 15 error parameters including the bias and scale factor error of the gyro and accelerometer.The precision of the method can also reach the precision of the systematic calibration method in stationary base.