Dynamic Airspace Design for Regional Air Mobility Operation in South Korea

Collaborative Research

Research Motivation

As the Regional Air Mobility (RAM) is a new air transportation system, there must be a need to establish a new airspace infrastructure that not only ensures efficient operations but shares the existing air traffic management system used for the civil aviation in a safe manner. In response to the new airspace network, the Korean government agencies have suggested using a structured airspace (e.g., corridor) at the early stage of development; however, the concept of a structured airspace may be limited to low-density operations. Besides, given that weather can potentially disrupt the RAM operations, it is unavoidable to accommodate the impact of convective weather in the process of airspace infrastructure establishment, which may result in creating dynamic geometry of airspace network possibly moving with convective weather. The main contribution of this research is to leverage a data-driven approach using machine learning techniques to dynamically establish the architecture of airspace network potentially utilized for upcoming RAM operations in South Korea.

Key Idea

This research proposes a novel methodology to construct the dynamic geometry of Korean airspace for RAM operations with the four modules:

1. The special use airspace identification module for taking into account airspace restriction areas

2. The partitioning-based clustering module for placing vertiport locations

3. The density-based clustering module for predicting areas of convective weather

4. The sampling-based Probabilistic Roadmap module for generating dynamic airspace network

Results

The results describe 1) special use airspace detection, 2) vertiport placements with aerial images, 3) an instance of convective weather polygons, and 4) dynamic airspace network generation.

Publication

• J. Kim and S. Kim, "Data-Driven Approach Toward Airspace Design for Regional Air Mobility Operations in Korea," Journal of Aerospace Information Systems, Vol. 20, No. 10, pp. 605-617, 2023. doi.org/10.2514/1.I011245