Space Object

Research on space objects focuses on improving the detection, tracking, and characterization of both known and unknown objects in orbit, primarily to address the growing problem of space debris and enable safe and efficient space operations. Current research employs a variety of techniques, including machine learning (e.g., convolutional neural networks, recurrent neural networks, and autoencoders), Structure from Motion algorithms, and neural radiance fields (NeRFs) for tasks such as object detection, pose estimation, 3D reconstruction, and orbit prediction. These advancements are crucial for enhancing space situational awareness, improving the safety of space missions, and facilitating the development of active debris removal and on-orbit servicing technologies.