Study On Algorithms Of Gravity Aided Inertial Navigation System

Inertial Navigation System is the main navigation system in underwater vehicle.Independent of the characteristics of passive Inertial Navigation System can guaranteeaircraft underwater hidden well and autonomy to meet the long-submergedunderwater aircraft requirements. However, Inertial Navigation System errorsaccumulate over time. After a long work it’s hard to ensure the accuracy of navigationpositioning. Navigation needs other ways aided inertial navigation to help. Gravityfield information acquisition’s method is simple. It’s a passive way to obtainindependent characteristics, widely used in aiding Inertial Navigation System. Gravityis a core part of the matching gravity aided Inertial Navigation System.In this thesis, the extraction point is matching gravity aided Inertial NavigationSystem to match the traditional iterative closest point algorithm (ICCP algorithm) forthe Inertial Navigation instruction nearest location and its corresponding contour pointfor a long time affecting the matching accuracy as well as Inertial Navigation Systemerror accumulating over time. The main question raised by the former method is tochange the structure of the Inertial Navigation instructions to extract the trajectorypoints and gradually reduce the scope of search for the nearest corresponding contourpoint extraction time. These two methods are to reduce the consumption of datasamples before when improving matching speed and efficiency. For InertialNavigation System error accumulation problem, the use of gravity matches through acombination of adaptive filtering algorithm Kalman filter on the key sources of errorestimate Inertial Navigation System, the feedback correction to compensate errors.Simulation results show that the improved algorithm improves the original methodICCP defects and improves the matching accuracy, estimation and compensation;improved Inertial Navigation System based on long-endurance navigation precisionposition results match Inertial Navigation System error well.Several works are carried on in this thesis:Based on the analysis of the advantages, disadvantages and limitations ofexisting matching algorithm, instead of traditional ICCP matching algorithm toextract the initial point of time sequence problem, use a single-point samplingmethod for sampling sequence matching algorithm to improve initial samplingsequence structure, reducing the consumption of match time, improving accuracy and real-time match data.Extract to match the initial point corresponding contour nearest point for a longtime and time-consuming calculations to gradually reduce the sampling points takenrecently contour corresponding point search method to reduce matching algorithms toextract data time and improve the matching efficiency.As a result of gravity match observations, use Kalman filtering to parameterestimation and compensation of Inertial Navigation System critical error tocorrect position errors of Inertial Navigation System. Experiments show that aftercompensation, Inertial Navigation System position errors are significantlyreduced before filtering. It shows the effectiveness of the proposed method.