Research On Star Identification Based On Shipborne Super-wide FOV Camera

Starlight Navigation is an autonomous navigation that uses the natural star of as the reference,obtains the coordinates of the star through passive observation,and determines the longitude,latitude and heading.Although satellite navigation systems such as GPS and Beidou have been continuously developed and widely used in recent years.Compared with satellite navigation,star navigation has the advantages of no electromagnetic interference,high reliability,low cost,and no error accumulation.Star identification is a key step in starlight navigation,and it is also a difficult point in the research of shipborne starlight navigation technology.The performance of starlight method will directly affect the accuracy and efficiency of the navigation system.In this paper,the practical requirements for navigation and positioning of shipbased platforms,especially large ships,are studied.A large-field starlight navigation system based on fisheye camera is designed.Aiming at the problems of large data,low recognition accuracy and low speed caused by the super-wide field of view(FOV),this paper focuses on a large FOV star identification method based on circular field of view segmentation.The method includes: optimizing the representation method of the star table data,proposing an improved star image feature extraction algorithm,and improving the star matching algorithm based on multiple triangles.Finally,it designs computer simulation experiments and marine observation experiments,the results of the experiments prove that it achieves the expected indicators and performance requirements,and then verifies the rationality and effectiveness of the star identification method.