A Metrics-based Method for Evaluating Corridors for Urban Air Mobility Operations

Xuan Jiang, Xin Peng, Vishwanath Bulusu, Cristian Poliziani, Gano Chatterji, Raja Sengupta2022 · DOI

TL;DR

This paper introduces a data-driven framework for evaluating Urban Air Mobility (UAM) corridors using real-world air traffic data instead of computationally expensive simulations. By applying spatial and temporal metrics to historical ADS-B data, the researchers provide a way for stakeholders to identify optimal "paths of least resistance" for air taxis that minimize conflicts with existing legacy aircraft.

What Problem Does This Paper Address?

As we move toward a future filled with electric vertical take-off and landing (eVTOL) aircraft, we need "highways in the sky"—known as air corridors—to manage traffic safely. Currently, evaluating where to put these corridors relies heavily on complex simulations and optimization models. These methods are often time-consuming, resource-intensive, and built on idealized assumptions that may not reflect reality.

Furthermore, different stakeholders have different priorities: a federal regulator cares about separation from legacy planes, a city planner cares about noise over residential areas, and an operator cares about transit efficiency. There is a clear need for a flexible, multi-objective metric system that can evaluate potential corridor designs using actual historical flight data.

Methods

The authors propose a "metrics-based selection" framework that treats air corridors as a series of rectangular cuboids. These cuboids are sized based on FAA regulations (e.g., 2,000 feet in height and 5 miles in width).

The core of their approach involves two models:

  1. Numerical Model: This assigns scores based on three primary metrics derived from historical traffic:
  2. $na$ (Traffic Volume): The total number of legacy aircraft entering the corridor space.
  3. $T$ (Occupancy Time): The total time legacy aircraft spend inside the proposed corridor.
  4. $nm$ (Non-transit Traffic): The number of aircraft staying longer than 4 minutes (indicating the space is a high-use area rather than just a crossing point). These are combined into a weighted score_total, allowing stakeholders to prioritize different factors.

  5. Heat Map Model: To find the best placement within a large region, the authors use a grid-based approach. They calculate a "cost" for every grid cell based on the inverse square of the distance ($1/d_{ij}^2$) between the grid center and legacy flight trajectories. Lower cost indicates a "quieter" area of the sky.

Datasets

The study utilizes a real-world dataset of ADS-B (Automatic Dependent Surveillance–Broadcast) track points:

  • Scope: 117,703 point measurements from January 30, 2022.
  • Geography: Within a 150-mile radius of Edwards Air Force Base in Los Angeles, California.
  • Scale: The data includes 3,506 unique aircraft with altitudes ranging from near sea level to over 14,000 feet.

Key Insights and Ideas

  • Simulation-Free Evaluation: The biggest "aha" moment is that you don't need a supercomputer to run a 24-hour traffic simulation to find a good corridor. By treating historical data as a "density field," you can identify gaps in existing traffic patterns statically.
  • Stakeholder Weighting: The scoring system is modular. If a regulator is worried about mid-air collisions, they can weight the "number of aircraft" metric higher. If an operator wants to avoid "loitering" legacy traffic that might cause delays, they can weight the "non-transit" metric higher.
  • Temporal Windows: Traffic isn't static. The researchers found that "cost" peaks significantly between 9:00 AM and 6:00 PM. This suggests that certain corridors might be viable for cargo at night but too congested for passenger air taxis during the day.
  • Orientation Sensitivity: The study tested eight different corridor orientations from a single center point. The variance in scores was high, proving that even a small shift in the angle of a corridor can dramatically change its safety profile.

Key Results

  • Optimal Corridor Identification: Out of eight tested orientations in the LA area, "Corridor 8" was identified as the clear winner, showing the lowest interaction with legacy traffic across all metrics.
  • Heat Map Precision: The heat map successfully identified a specific grid location (Location 9) with a minimum cost of 53.002, compared to higher-traffic areas nearby that peaked at 56.133.
  • Validation of Metrics: The variance analysis (Table III) confirmed that the proposed metrics are sensitive enough to distinguish between different corridor designs, making them a valid tool for decision-making.

Limitations and Future Work

While the paper provides a robust framework for airspace safety, the authors acknowledge several areas for expansion:

  • Ground Risk: The current model doesn't account for what's under the corridor. Future versions need to integrate population density data to minimize risk to people on the ground.
  • Environmental Impact: Noise and emissions data should be layered in to satisfy municipal planning requirements.
  • Multi-Layer Corridors: The study focused on single-layer corridors. Future research will look at "stacked" corridors to handle higher volumes of UAM traffic.
  • Dynamic Data: Moving from historical "static" data to real-time or more complex urban air traffic data would improve the model's predictive power.

Summary updated

Abstract

Air corridor evaluation is important for the planning and organization of urban air mobility operations. However, aviation metrics are typically evaluated by simulation and optimization, which can be redundant and bulky or heavily rely on unrealistic assumptions. We propose to explore metrics to evaluate urban air corridor design in congested areas that (1) are based on traffic data (i.e., without the need for simulation), and (2) serve different temporal and spatial (i.e. multi-objective) purposes. The metrics will account for different stakeholder interests in the evaluation of the corridor designs. For this study, the corridor is described as a bigger cube combined with several small cubes based on the vertical and lateral distance thresholds derived from FAA regulations. Aircraft time, the number of aircraft within the cube within a certain period and location slices of the corridor are used to evaluate the proposed air corridors. This paper discusses methods for choosing the best corridors from a given set as well as the best route for a given region. The proposed methodology is demonstrated with a case study based on air traffic data within 150 miles of Edwards Air Force Base in Los Angeles, California. Experiments show that air traffic data can help stakeholders make both spatial and temporal decisions.
BibTeX
@misc{jiang2022metrics,
  author = {Xuan Jiang and Xin Peng and Vishwanath Bulusu and Cristian Poliziani and Gano Chatterji and Raja Sengupta},
  title = {{A Metrics-based Method for Evaluating Corridors for Urban Air Mobility Operations}},
  year = {2022},
  doi = {10.31219/osf.io/fqu3k},
}
Figure