Multi-agent learning for data-driven air traffic management applications

Chuhao Deng, Hong-Cheol Choi, Hyunsang Park and Inseok Hwang

Journal of Air Transport Management, 2025, vol. 128, issue C

Abstract: Research in developing data-driven models for Air Traffic Management (ATM) has gained tremendous interest in recent years. However, data-driven models are known to have long training time and require large datasets to achieve good performance, and the majority of proposed data-driven models ignores ATM system’s multi-agent characteristic. To fill the research gaps, this paper proposes a Multi-Agent Bidirectional Encoder Representations from Transformers (MA-BERT) model, which fully considers the multi-agent characteristic of the ATM system and outputs results based on all agents in the airspace. Additionally, compared to most data-driven models that are designed for a single application, the proposed MA-BERT’s encoder architecture enables it to be pre-trained through a self-supervised method and fine-tuned for a variety of data-driven ATM applications, saving a substantial amount of training time and data usage. The proposed MA-BERT is tested and compared with other widely used models using the Automatic Dependent Surveillance-Broadcast (ADS-B) data recorded in three airports in South Korea in 2019. The results show that MA-BERT can achieve much better performance than the comparison models, and by pre-training MA-BERT on a large dataset from a major airport and then fine-tuning it to other airports and applications, a significant amount of the training time can be saved. For newly adopted procedures and constructed airports where no historical data is available, the results show that the pre-trained MA-BERT can achieve high performance by updating regularly with small amount of data.

Keywords: Air traffic management; Trajectory prediction; ETA prediction; Multi-agent system; Transfer learning; Machine learning (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0969699725001061
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:jaitra:v:128:y:2025:i:c:s0969699725001061

DOI: 10.1016/j.jairtraman.2025.102843

Access Statistics for this article

Journal of Air Transport Management is currently edited by Anne Graham

More articles in Journal of Air Transport Management from Elsevier
Bibliographic data for series maintained by Catherine Liu ().