NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
VFR Trajectory Forecasting using Deep Generative Model for Autonomous Airspace OperationsTo enable the airspace integration of autonomous operations, such as uncrewed aircraft conducting cargo deliveries, there is a need to forecast the positions of the surrounding traffic with which they may interact. This paper focuses on forecasting Visual Flight Rules traffic, a significant source of uncertainty and risk in the airspace, especially around small regional airports, due to the unplanned and often untracked nature of such flights. A deep generative model is developed, trained on historical traffic data at example towered and non-towered airports, and used to predict flight trajectories. Experimental results are presented comparing the performance of variational autoencoder and classical machine learning forecasting when applied to both the towered and non-towered airports over varying time horizons. The results show the advantages of the variational autoencoder in producing accurate probabilistic forecasts over varying time horizons.
Document ID
20240008243
Acquisition Source
Ames Research Center
Document Type
Conference Paper
Authors
Aastha Acharya
(Metis Technology Solutions, Inc. Albuquerque, NM)
Vishwanath Bulusu
(Crown Consulting, Inc Arlington, VA)
Husni Idris
(Ames Research Center Mountain View, United States)
Date Acquired
June 27, 2024
Subject Category
Air Transportation and Safety
Meeting Information
Meeting: 43rd Digital Avionics Systems Conference (DASC)
Location: San Diego, CA
Country: US
Start Date: September 29, 2024
End Date: October 3, 2024
Sponsors: Institute of Electrical and Electronics Engineers, American Institute of Aeronautics and Astronautics, Digital Avionics Technical Committee
Funding Number(s)
PROJECT: 629660
CONTRACT_GRANT: 80ARC018D0008
CONTRACT_GRANT: NNA16BD14C
Distribution Limits
Public
Copyright
Portions of document may include copyright protected material.
Technical Review
NASA Technical Management
Keywords
uncrewed aircraft
autonomous operations
autonomy
machine learning
trajectory forecasting
trajectory prediction
generative model
deep learning
No Preview Available