Multi-Agent Based Transfer Learning for Data-Driven Air Traffic Applications

TL;DR

This paper introduces a multi-agent transfer learning model based on MA-BERT for air traffic management, significantly reducing training time and enabling high performance with limited data across different airports.

Contribution

It proposes a novel multi-agent BERT model and a transfer learning framework tailored for ATM, addressing data scarcity and training efficiency issues.

Findings

Pre-trained MA-BERT reduces training time substantially.

High performance achieved with minimal data for new procedures.

Effective across multiple airports and data scenarios.

Abstract

Research in developing data-driven models for Air Traffic Management (ATM) has gained a tremendous interest in recent years. However, data-driven models are known to have long training time and require large datasets to achieve good performance. To address the two issues, this paper proposes a Multi-Agent Bidirectional Encoder Representations from Transformers (MA-BERT) model that fully considers the multi-agent characteristic of the ATM system and learns air traffic controllers' decisions, and a pre-training and fine-tuning transfer learning framework. By pre-training the MA-BERT on a large dataset from a major airport and then fine-tuning it to other airports and specific air traffic applications, a large amount of the total training time can be saved. In addition, for newly adopted procedures and constructed airports where no historical data is available, this paper shows that the pre-trained MA-BERT can achieve high performance by updating regularly with little data. The proposed transfer learning framework and MA-BERT are tested with the automatic dependent surveillance-broadcast data recorded in 3 airports in South Korea in 2019.