Construction of Computer Vision Guidance System for Spacecraft Orbit Based on Artificial Intelligence Algorithm

Accurate navigation and positioning of spacecraft is the key factor to ensure the successful implementation of its mission. In order to cope with the complex and changeable space environment, a computer vision guidance system for spacecraft orbit based on artificial intelligence (Al) algorithm is proposed in this study. Based on the deep learning algorithm, the system aims to realize an efficient and robust navigation system by processing and analyzing the visual information of spacecraft orbit in real time. In this study, neural network can approximate any nonlinear function, the relative motion state of the spacecraft at the current moment and the remaining time of the pursuit game are taken as inputs, and the maneuvering acceleration of the spacecraft during the pursuit process is taken as output. The internal relationship contained in these data is studied by using deep neural network, and an online generation method of maneuvering strategy of the spacecraft pursuit game with fixed stay based on deep neural network is obtained. By training on large-scale data sets, we demonstrate the generalization ability of the deep learning model in unknown orbit scenes. The good generalization of the model makes the system have the ability to adapt to different tasks and environments, which provides strong support for the reliable implementation of practical tasks.