FLIGHT DESTINATION PREDICTION BASED ON BAYESIAN UPDATING USING ADS-B DATA
Loading...
Authors
Lee, Minkyu
Subjects
ADS-B
predict
destination
trajectory
Bayesian
model
phase
predict
destination
trajectory
Bayesian
model
phase
Advisors
Koyak, Robert A.
Kim, Suhwan, Korea National Defense University
Date of Issue
2020-03
Date
Publisher
Monterey, CA; Naval Postgraduate School
Language
Abstract
The constant increase in air demand worldwide is expected to lead to a surge in traffic control, which has raised questions about air traffic safety and air traffic control efficiency. The objective of this thesis is to create a model that effectively utilizes a small amount of data to predict the destination of an aircraft probabilistically. The model also can be used to predict anomalous situations. The destination forecasting model is based on Bayesian updating using Automatic Dependent Surveillance-Broadcast (ADS-B) data. ADS-B data contains aircraft coordinates, heading, speed, ID, time, and squawk. The Bayesian model performance is measured to the extent that the probability of a destination forecast for aircraft originating from Dublin converges to 12 destinations. We find that the model predictive accuracy of our model exceeds 75% where convergence to a destination is defined as having a probability greater than .95 for three consecutive iterations. These probabilistic destination predictions not only contribute to commercial air traffic control management but also may be useful to air surveillance activities such as anomaly detection.
Type
Thesis
Description
Series/Report No
Department
Operations Research (OR)
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Approved for public release; distribution is unlimited.
Rights
Copyright is reserved by the copyright owner.