A Voice-Communication Augmented Simulation (VCAS) Framework for Aircraft Trajectory Simulation

摘要

The near-terminal operations of the aircraft highly rely on the ATC command. Accurate prediction for aircraft trajectory can help the safety operation in the National Airspace System (NAS). This paper presents a novel simulation framework that considers the voice communication between the pilot and Air Traffic Controllers (ATC) to assist the trajectory simulations. The simulation is based on NATS, a newly developed open access trajectory simulation platform. The framework utilizes a state-of-the-art voice-to-text engine specifically tuned for ATC environment to translate the audio data into text. A Natural Language Understanding (NLU) processes the text and extract key information such as aircraft call sign, related operation command etc. This information can them be used to update the NATS simulation code, resulting in a simulated trajectory augmented with voice commands. The framework will be referred as Voice-Communication Augmented Simulation (VCAS). The current VCAS can utilize the command in leveling change and runway assignments. The proposed method is validated with real trajectory data from Sherlock data warehouse.