Cissi Y. Lin
Department of Space Science & Engineering
Center for Astronautical Physics & Engineering
National Central University
In recent years, the global space race has intensified, leading to a rapid increase in the number of on-orbit satellites and debris. The risk of collisions between space objects has become a major challenge threatening global space security. Our goal of this study is to make develop light-weight AI platform of high performance and low power consumption for monitoring and responding to space situational awareness related issues. With a particular focus on YOLO-based models, we address the critical challenge of detecting and classifying orbital objects with limited training data and computational resources. We evaluate data augmentation and preprocessing strategies that will further enrich the training dataset, improving the system’s ability to generalize across synthetic and real-world imagery.
