| Due to the rich oil and gas resources,dominant strategic position and the advantage in shortening the voyage of carrier,polar regions have attracted worldwide attention.The loss of glaciers caused by global warming provides the feasibility for ship sailing in polar regions.Strapdown Inertial Navigation System(SINS),which is not affected by the special climatic conditions and geographic position in polar regions,can output the attitude,velocity and position of the ship independently,and has been selected as the first choice for shipborne polar navigation.In order to overcome the difficulty of north-oriented SINS mechanizaiton in determining the position and yaw in polar regions,grid SINS mechanization is proposed and taken as the reliable mechanization for polar navigation.Due to the measurement errors of inertial sensors and other error sources,the errors in SINS manifest as oscillating error and accumulated error,and the errors drift with time dramatically.In this paper,aiming at shipborne grid SINS,the theory and method of external level damping and comprehensive calibration are studied to restrain the Schuler oscillating error and accumulated error.The main work of this paper includes:1.The external level damping algorithm is an effective method to restrain the Schuler oscillating error in SINS.The overshoot occurs when SINS switches from non-damping working state to external level damping working state.An external level damping algorithm based on Kalman filter is proposed to solve this problem.In the proposed algorithm,the Kalman filter is designed based on the error equation of shipborne grid SINS,with the velocity given by Doppler Velocity Log(DVL),the level attitude errors are estimated by Kalman filter and the Schuler oscillating error is dampened by compensating the level attitude errors through feedback calibration.Meanwhile,the state switching criteria is designed based on the changing features of observations which aims to remove the burdens of manual switch.By starting the feedback calibration when the estimations of level attitude errors are stable,the proposed algorithm reduces the oveshoot and shortens the damping adjustment time.2.In nonsensitive regions,the shipborne SINS usually navigates in external level damping working state.The comprehensive calibration scheme based on intermittent calibration information is an effective method to suppress the accumulated errors in damping SINS.However,the motion state of ship is restricted in traditional scheme and the three-point calibration method cannot be applied in SINS,meanwhile the existing P equation deduced in geographic frame does not match the grid SINS.Aiming at these problems,first,the P equation used in comprehensive calibration is deduced in grid frame.Next,in designing the comprehensive calibration schemes,the OEPQ frame is replaced by inertial frame to overcome the decfects in traditional scheme.Finally,on the basis of P equation and ?equation,the two-point and three-point comprehensive calibration schemes are designed.Since the gyroscope drift is accurately estimated and well compensated,the accumulated errors in SINS are restrained,and problem that the three-point comprehensive calibration scheme cannot be applied in SINS is solved.3.When the ship sails in sensitive region,SINS navigates in non-damping state.In order to suppress the accumulated errors in SINS,with considering reducing the time of using DVL and improving the estimation accuracy of gyroscope drift,the non-damping two-point and three-point comprehensive calibration schemes are designed.First,the DVL-aided joint parameter estimation method is proposed to estimate the attitude and acceleormeter biases.The parrameter estimation method is used to estimate and compensate the accelerometer biases before comprehensive calibraition,meanwhile the level attitude errors in calibration are also estimated by the method.Then the P equation in which the level attitude errors are reserved is derived in grid frame.Finally,on the basis of P equation,? equation and the deduced joint parameter estimation method,the non-damping two-point and three-point comprehensive calibration schemes are designed to estimate the gyroscope drift.Since the gyroscope drift is accurately estimated and well compensated,the accumulated errors in SINS are restrained.As the accelerometer biases are compensated before comprehensive calibration,the level attitude steady-state errors are eliminated,which improves the estimation accuracy of gyroscope drift.Meanwhile,since the parameter estimation method spends a short time to estimate accelerometer biases and level attitude,the proposed schemes reduce the time of using DVL output,which maintains the navigation autonomy of the ship to the greatest extent.4.With continuous calibration information,two types of optimal comprehensive calibration schemes are proposed.The first optimal comprehensive calibration scheme is realized with the continuous position and yaw which aims to estimate the gyroscope constant drifts in short time with high accuracy.The gyroscope constant drifts are extended in the state variables.The measurement equaiton of Kalman filter is established based on the P equation in damping state,and the state equation of Kalman filter is established based on the ? equation.With the external position and yaw,the gyroscope constant drifts are estimated through Kalman filter in short time wih high accuracy,and the navigation errors are restrained effectively.The second optimal comprehensive calibration scheme is realized with the continuous position and velocity.The double calculating circuits are introduced to eliminate the restriction of existing scheme on the working state of SINS.The first circuit,named navigation circuit,can work in any working state and outputs the navigation parameters of the ship;the second circuit,named calibration circuit,works in non-damping state.In second circuit,the Kalman filter is established based on the error equation of shipborne grid SINS,and the gyroscope constants drifts and navigation parameter errors are estimated by Kalman filter with the external position and velocity.When the calibration information terminates,the gyroscope drifts are compensated and the navigation parameters are reset based on the results of observability analysis,which bounds the accumulated errors in navigation circuit. |
PDF Full Text Request