Configuration Estimation from Position Data of Civil Jet Aircraft for Noise Mapping

This study presents a novel machine learning approach for estimating the flap and landing gear configurations of civil jet aircraft based on position data. Optimized for noise calculations, this method simplifies the estimation process by grouping various configuration states, thereby reducing the potential for estimation errors and focusing on configurations significant for noise impact assessments. Validation on holdout data demonstrates the method’s high accuracy and potential for improving aircraft noise predictions. When integrated into the aircraft noise model sonAIR, the inclusion of estimated configuration settings significantly improves noise calculation accuracy compared to methods that account for aircraft configuration only implicitly. Especially for wide-body aircraft during arrival, we could improve the calculation accuracy in noise-exposure-relevant areas by several decibels.