Orbit Model

Orbit modeling focuses on accurately predicting the trajectories of space objects, crucial for space situational awareness (SSA) and collision avoidance. Current research emphasizes using machine learning, particularly deep learning architectures like transformers and neural networks integrated with traditional methods such as SGP4, to improve prediction accuracy and efficiency, especially when dealing with non-conservative forces and limited data. This improved modeling is vital for enhancing space safety and security, enabling more effective space traffic management and debris mitigation strategies. Furthermore, distributed and federated learning approaches are being explored to address data privacy and bandwidth limitations in large satellite constellations.