Machine-Learning-Based Real-Time Trajectory Prediction of Store Release from Cavity

What is it about?

We study real-time prediction of a store’s path after release from an internal aircraft bay, where unsteady cavity flow makes trajectories difficult to forecast. We develop a data-driven reduced-order model trained on 2D CFD outputs, using clustering, dimensionality reduction, and regression to generalize to new cases. The model predicts horizontal/vertical positions and orientation with moderate to satisfactory accuracy relative to CFD. Performance degrades on finer vertical-motion details (higher-order modes), but results indicate strong potential for operational, real-time use.

Featured Image