A Map-Matching Algorithm for Ground Movement Trajectory Representation using A-SMGCS Data

• Published in: 2020 International Conference on Artificial Intelligence and Data Analytics for Air Transportation (AIDA-AT) pp. 1 - 8
• Main Authors: Tran, Thanh-Nam, Pham, Duc-Thinh, Alam, Sameer
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.02.2020
• Subjects: A-SMGCS Data; Aircraft; airport graph; airport surface movement; Airports; Clustering algorithms; data analytics; map-matching; Meters; Prediction algorithms; Tracking; Trajectory
• DOI: 10.1109/AIDA-AT48540.2020.9049181

Increasing availability of air traffic data has opened new opportunities for better understanding of Air Traffic Management (ATM) system. At Airport-Air side, A-SMGCS (Ad-vanced Surface Movement Guidance & Control System) data may provide useful insights to improve efficiency and safety of airport operations by understanding traffic patterns, taxiway usage, ground speed profiles and any anomaly behaviour. However, A-SMGCS data comes from the fusion of several sensors such as MLAT, ADS-B and SMR. This leads to high and variable noise, missing data values, and temporal and spatial misalignment. In this study, we proposed a new and simplified representation of ground movement trajectories using a map-matching algorithm applied on A-SMGCS data. The proposed approach not only overcomes above mentioned issues of data, but also takes into consideration airport specific operational constraints. The algorithm shows a good matching results with mean percentage error of approximate 8.13%. The matching trajectories and sequences of nodes in resulting graph, supports a variety of analysis about airport operations. To show the effectiveness of proposed approach, we performed some analysis such as traffic patterns, taxi-way usages, speed profiling and anomaly detection, using one month of A-SMGCS data at Singapore Changi Airport.