Research On Deeply Coupled Of Laser Gyro And Star Sensor

In the field of modern navigation,Strapdown Inertial Navigation System(SINS),which possesses the characteristics of complete navigation information,strong autonomy,high accuracy within short time,pretty stability,fast updating frequency of data,good adaptability to the environment,etc.,can provide various precise navigation information for the carriers,but its error will be accumulated with time;Celestial Navigation System(CNS),whose navigation core is star sensor(SS),can reflect the attitude information of carriers with no relation with time,however,its updating frequency is low and it is easy to be affected by the environment.Therefore,the combination of SINS and SS can make up their weakness for each other efficiently and develop their superiority in order to realize the navigation of perfect performance.Recently,the measuring accuracy of SS is related with the motion situation,and the SS error under dynamic conditions is a bottleneck of limiting the development of SINS/CNS navigation accuracy and the application of SINS/CNS.Aimed at this problem,based on the analysis on the error character of SS,the attitude-correlated frames(ACF)approach is presented.Furthermore,the device-level deeply integrated navigation methods with laser gyro and SS can be well studied on this basis.As a result,the improvement of integrated attitude accuracy will be true.The primary content of this dissertation is as follows:1)The fundamental theories of SS are introduced detailedly first,and the imaging model of SS is proposed.Meanwhile,the measuring principle of SS attitude and the SS error under dynamic conditions are analyzed.On one hand,the effect of the centroiding noise on measuring attitude is studied through simulation and it can be found that the centroiding noise influences more the cross boresight attitude angle error than the boresight error.On the other hand,through the research on the dynamic performance of SS and the effect of angle motion,the principle of dynamic error of SS can be made out,and then two approaches are introduced briefly to solve the motion-blurred star image of single frame.2)The dynamic accuracy of SS can be improved based on the attitude-correlated frame(ACF)method.The principle and the effect on measuring accuracy of correlating number are introduced firstly.The results of simulations and experiments demonstrate that the suppression effect on the random noise of ACF approach can be enhanced with the increase of the number of attitude-correlated frames and these results can verify the availability of ACF.Then,through simulation,it can be obtained that gyro bias,random walking noise,and correlating frame number have effect on the improvement of ACF.When the ACF approach are used with gyros of different characteristics,the maximum number of attitude-correlated frames will exist to best strengthen the performance of SS under dynamic conditions.3)Depending on the basis that the SS error can be suppressed through ACF approach and the high-accuracy attitude measuring information can be obtained.Therefore,the device-level deeply integrated navigation methods utilizing the characteristics of the SS to eliminate gyro error via comparing the errors of both SS and gyro at the same measuring frame are proposed.Depending on this theory,two kinds of device-level deeply integrated navigation methods of laser gyro and SS are come up with in this paper.The method Ⅰ algorithm estimates the gyro bias using the adjacent frames,and the result is used for the feedback to ACF correlating matrix.Through real-timely updating,the attitude error can be corrected.The method Ⅱ algorithm utilizes the accumulated results of SS from the initial frame to present frame to estimate both gyro bias and attitude error.Through the accumulation of time,it will be more beneficial to compensate the gyro error.After those simulations on the effectiveness of the two algorithms,it can be found that the accuracy of navigation can be improved,especially method Ⅱ.Finally,the simulation results indicate that the bias error of SS has more effect than the white noise.