Airport Taxi Time Prediction and Alerting: A Convolutional Neural\n Network Approach

2021/11/17 by Erik Vargo, Vargo, Erik, Alex Tien +3
Economics, Econometrics and Finance · Engineering · #Aerospace and Aviation Technology #Air Traffic Management and Optimization #Artificial Intelligence (cs.AI) #Aviation Industry Analysis and Trends #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2111.09139

openalex publication_date 2021/11/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper proposes a novel approach to predict and determine whether the\naverage taxi- out time at an airport will exceed a pre-defined threshold within\nthe next hour of operations. Prior work in this domain has focused exclusively\non predicting taxi-out times on a flight-by-flight basis, which requires\nsignificant efforts and data on modeling taxiing activities from gates to\nrunways. Learning directly from surface radar information with minimal\nprocessing, a computer vision-based model is proposed that incorporates airport\nsurface data in such a way that adaptation-specific information (e.g., runway\nconfiguration, the state of aircraft in the taxiing process) is inferred\nimplicitly and automatically by Artificial Intelligence (AI).\n

Related

• Flight Delay Prediction using Airport Situational Awareness Map
• Aircraft Landing Time Prediction with Deep Learning on Trajectory Images
• Model Generalization in Arrival Runway Occupancy Time Prediction by Feature Equivalences
• On the Statistics and Predictability of Go-Arounds
• Artificial Neural Networks Applied to Taxi Destination Prediction