Research on the Star Recognition Method Based on Star Trackers
|School||Harbin Institute of Technology|
|Course||Mechanical and Electronic Engineering|
|Keywords||Star sensor Star recognition Triangle algorithm Star track identification algorithm|
The high-precision attitude measurement system is an important guarantee for the various types of spacecraft to complete a specific space mission . Access to various spacecraft flight attitude information system are collectively referred to as the attitude measurement system or attitude sensors . According to different principles and measurement methods , the researchers have developed a variety of attitude sensors : earth sensor and sun sensor , magnetometer , gyroscope and star sensor . Compared to other types of attitude sensors , star sensor has the advantages of high accuracy , light weight , low power consumption , no drift and ways , and at the same time a new generation of star sensor and inertial gyro , like autonomous navigation capabilities , enabling no gyro spacecraft attitude determination system is possible. The star sensor is an application promising attitude sensors . Star identification software algorithm is one of the key technologies in the star sensor . This thesis proceed from the overall structure of star sensor software algorithms , system analysis and implementation of the various parts . Star recognition algorithm based on in-depth study , in order to achieve a substantial increase in the star sensor purpose of recognition success rate and frequency of updates . In this thesis, complete the following tasks : ( 1 ) analysis and achieve recognition mode of the celestial sphere , improved and perfected the triangle recognition algorithm based on Star recognition , in the case of the recognition time without a significant increase , a substantial increase in The success rate of 98.5 % . (2) recognition mode on the basis of the entire celestial sphere , to achieve the star charts extraction based on satellite tracking mode and star identification ( recognition time is approximately 0.04 seconds) greatly improve the recognition success rate and speed of operation , while the method significantly improves anti-interference ability and stability of the software system .