A XGBoost-Based Movement Prediction System Using ADS-B Data

The Automatic Dependent Surveillance-Broadcast (ADS-B) technology is a cost-effective airplane surveillance system, which has many applications. In this paper, we present an eXtreme Gradient Boosting (XGBoost)-based flight prediction model and analyze the differences from actual flight paths by evaluating various hyperparameter combinations using ADS-B data. We utilized R \(^2\) and MAE to find the optimal hyperparameters for longitude and latitude, respectively. The evaluation results confirm that the proposed model can accurately predict aircraft movements using ADS-B data.