A Finite Automata Based Method for Autonomous Star Identification

A star identification algorithm Based on finite automata is proposed in this paper. Firstly, each star is characterized by a well-defined pattern that can be determined by its surrounding stars and generated with a virtual circle. Then, the star pattern is coded into strings and a guide catalog is also constructed according to the generating process of the star pattern. Finally, a string matching method Based on finite automata is introduced to the star identification process to find the closest match between observed patterns and catalog patterns. The simulation result shows that the algorithm performs better than the grid algorithm in terms of robustness and stability, and provides a good strategy to solve false stars problem.