Research On SINS/CNS Integrated Navigation Technique Of Aerospace Vehicle

The special flight environment and characteristic of Aerospace Vehicle impose particular high requirements on the autonomy, reliability and accuracy of its navigation system. Characterized by a long flight time and a wide flight range, in-orbit flight segment is a crucial phase of Aerospace Vehicle for implementing various tasks and flight experiments. Thereby, ensuring the navigation information to be high-reliable, high-autonomic and high-accurate for this phase is an absolute necessity. At present, relying solely on inertial navigation system cannot meet the accuracy requirement of this phase. INS/GPS integrated navigation system, on the other hand, though provides highly accurate navigation information, cannot be used in actual operation, because of its limited autonomy. As a result, it is of great significance to develop a completely independent navigation system for space vehicle. Taking its motion characteristics of in-orbit flight segment into account, a study of key INS/ CNS integrated navigation technology is carried out in this thesis in the hope of being able to provide a highly accurate INS/CNS integrated navigation programs and algorithms suitable for Aerospace Vehicle’s in-orbit flight. Thus, theoretical foundation for creating the independent navigation system of high accuracy would be laid.During the Aerospace Vehicle’s in-orbit flight segment, guide stars observed by celestial sensor change over time, affecting the performance of celestial positioning. Considering this, an error model based on celestial positioning principle is established in this thesis. Based on this, various types of factors affecting the celestial positioning errors are analyzed, and the error weighting coefficient for assessment is provided. At the time, laws of the effects of stellar geometry on celestial positioning performance are studied. Then based on the results, design a star selection algorithm for celestial positioning, which would improve its performance remarkablely.Given that alternation of guide stars also have influence on astronomical attitude determination performance, an error model based on basic astronomical attitude determination principles is established in this thesis by applying covariance analysis, which provide error weighting coefficients for assessment. Adopting the method of tri-star attitude determination, laws of effects of star tetrahedron volume on celestial attitude determination performance is chiefly studied. Then based on the results, design a star selection algorithm for celestial attitude determination, which would enhance its performance effectively.On the basis of the studies mentioned above, an INS/CNS integrated navigation architecture based on gyro error estimation is outlined to achieve high accuracy in the correction of inertial navigation systems thorough a comprehensive utilization of celestial navigation information. A gyro error correction method based on INS/CNS attitude integrated navigation under inertial coordinate is devised, and gyro drift errors it estimated are feedback to the work process of inertial navigation system in geographic coordinate. Then an INS/CNS integrated positioning navigation module suitable for time-varying noise is designed by introducing celestial positioning information into the devised module. By using the celestial positioning information, errors of inertial navigation system in geographic coordinates can be further rectified. Thereby a comprehensive revise of navigation system can be achieved.Since it is hard to obtain an accurate error model of actual navigation system in Aerospace Vehicle’s high-dynamic flight, a further study of robust filtering for INS/CNS integrated navigation algorithm of Aerospace Vehicle is carried out in this thesis. By applicating H∞ robust filtering, an INS/CNS integrated navigation system based on H∞ filter is established, and the filter’s performance when there is difference between error model of actual navigation system and the filter’s model is studied. In the meantime, the influence of various robust parameters on navigation performance through numerical simulation is analyzed.On the theoretical basis of INS/CNS navigation algorithm, an INS/CNS integrated navigation simulation platform based on STK/Matlab is created in this thesis. Therefore, provides a favorable platform to test the INS/CNS integrated navigation technology for aerospace vehicle by simulating its in-orbit flight path, inertial navigation systems, astronomical navigation subsystem etc.