Marine Multi-INS Cooperative Positioning And Error Parameter Estimation

In order to meet the demand of modern warships on long-endurance and high-precision navigation,extensive research has been carried out on rotational inertial navigation system(RINS)at home and abroad.At the level of existing inertial sensor,the introduction of rotation modulation technology greatly improves the accuracy of the inertial navigation system(INS).And the RINS has been widely used in various types of warships.In order to ensure the reliability,the warships are generally equipped with multiple different types of INS,such as the electrostatic suspend gyroscope INS,the platform INS,the fiber optic gyroscope INS,the single-axis RINS and the dual-axis RINS.In the redundant configuration,only one system is generally used as the master INS,while the rest are used as the hot-backup systems.There is a lack of information fusion.In recent years,with combining several sets of INS,the researchers try to improve the overall reliability and the accuracy of the systems through information fusion.The problem is that there is no effective method and model for information fusion to optimize the INS information.To solve these problems,this paper mainly completed the following work:1.A novel joint error parameters estimation method based on joint rotation modulation is proposed for the redundant INSs.The attitude error difference,the velocity error difference and the position error difference between the differenrt INSs are taken as the error states,and the navigation parameter differences are used as observations.If the error characteristics of different INSs appear to be locally different or complementary to each other,some(or all)of the gyro drifts and aceelerometer biases can be estimated.On one hand,the eatimates of these inertial sensor biases can be used to monitor the working state of the INS;on the other hand,by means of output correction,the position error caused by these gyro drifts and accelerometer biases can be compensated to a certain extent,which improves the accuracy and the reliability of the redundant systems at the same time.2.A cooperative positioning method based on grid frame is proposed for the redundant INSs.A joint error states Kalman filter is redesigned to meet the demand of warships on high-latitude(polar-region)sailing.The transformation relationship of joint error states between the geographic frame and the grid frame is also established.The redesigned joint error states Kalman filter is with global applicability,and without longitude and true North orientation calculation problem caused by the meridian convergence in the polar region.3.An online performance evaluation method of RLG RINS based on joint rotation modulation is proposed.Two parallel computing joint error states Kalman filters are constructed between a single-axis RINS and two dual-axis RINSs.These filters work to estimate the single-axis RINS azimuth gyro constant drift,and the standard deviation of that is used as a criterion to pick out the better dual-axis RINS with less random error.The semi-physical simulations and the experimental results show that the RLG random error difference of the dual-axis RINS can be distinguished at a level of 10% noise.4.A novel transfer alignment method based on joint rotation modulation is proposed.In the case of redundant single-axis RINS configuration,the joint error states Kalman filter is designed to estimate the horizontal inertial sensor biases,whereby the Schuler oscillation error and the saw-tooth error in the velocity output of single-axis RINS can be predicted and compensated.With application to slave INS transfer alignment,the compensated effectiveness of the velocity error of single-axis RINS is verified.5.The dynamic model of laser gyro is estabilished,and the mechanism of laser gyro g-sensitive error is analyzed in detail,whereby the g-sensitive misalignments for RLG-triad is deduced.When the angular rate is present,the equivalent gyro drift caused by g-sensitive misalignment will occur.The influence of g-sensitive misalignment is great in the linear vibration and angular oscillation coexisting environments.With optimized observation scheme,a novel calibration method based on linear vibration is proposed to estimate the g-sensitive error parameters of RLG-triad.And the linear vibration experiments are performed on the INS to verify the proposed method.The compensation for the g-sensitive misalignment of RLG-triad is of importance to improve the accuracy of shipborne INS in mechanical environment.