Star pattern recognition method based on neural network

Star sensor is an avionics instrument used toprovide the absolute 3-axis attitude of a spacecraft by utiliz-ing star observations. The key function is to recognize theobserved stars by comparing them with the reference cata-logue. Autonomous star pattern recognition requires thatsimilar patterns can be distinguished from each other with a small training set. Therefore, a new method based on neural network technology is proposed and a recognition systemcontaining parallel backpropagation (BP) multi-subnets isdesigned. The simulation results show that the method per-forms much better than traditional algorithms and the pro-posed system can achieve both higher recognition accuracyand faster recognition speed.