Dominic Bruno, Christopher Yeung, Justin Gmys, Binh Tran
Space Weather Challenge is an MIT ARCLab/STORM-AI repository for predicting space-weather-driven atmospheric density changes in low Earth orbit.
It includes the team’s Phase 1 competition submission, training and preprocessing code, Codabench submission files, development tools, DVC-managed data setup, and troubleshooting notes. The model supports nowcasting and forecasting density variation, which helps with satellite tracking and orbit resilience analysis.