Проблемні питання розробки систем динамічної генерації сценаріїв для сучасних диспетчерських тренажерів

Abstract

This study addresses automating scenario creation for air traffic control simulators within the CBTA framework. Manual modeling limits training variability and causes a "learned scenario" effect; therefore, a data-driven approach is proposed to automatically adapt task complexity to the cadet's level. The paper analyzes technical barriers, including data incompleteness and the complexity of modeling dynamic ATC behavior. The findings offer mechanisms to enhance training realism while optimizing the workload for instructors.