Aircraft trajectory prediction and risk assessment using bayesian updating

Yuhao Wang, Yutian Pang, Yongming Liu, Parikshit Dutta, Bong Jun Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Scopus citations


Flight trajectory prediction is crucial in maintaining the safety and predicting accidents in the National Airspace System (NAS). The reported work used Bayesian updating to achieve flight trajectory prediction and real-time risk assessment in the NAS. The trajectory simulation is done using NATS, a novel flights simulation platform. The model can consider multiple sources of uncertainties such as weather, human performance etc. Through Bayesian updating, the uncertainty in the model can be reduced given observable quantities. In this article, the Bayesian framework in updating model parameter through observation is introduced. The NATS simulation for a real accident scenario at SFO airport will be presented. In the presented framework, the risk probability is updated continuously using the aircraft location tracking information. The accident can be predicted well before it happens. A criterion for assessing the risk probability is developed under the NATS platform. The risk probability is evaluated based on the separation between aircrafts. It can work as a computer-aided algorithm for Air Traffic Management (ATM) aiming to help the ATC operator in preventing potential accidents.

Original languageEnglish (US)
Title of host publicationAIAA Aviation 2019 Forum
PublisherAmerican Institute of Aeronautics and Astronautics Inc, AIAA
ISBN (Print)9781624105890
StatePublished - 2019
EventAIAA Aviation 2019 Forum - Dallas, United States
Duration: Jun 17 2019Jun 21 2019

Publication series

NameAIAA Aviation 2019 Forum


ConferenceAIAA Aviation 2019 Forum
Country/TerritoryUnited States

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Aerospace Engineering


Dive into the research topics of 'Aircraft trajectory prediction and risk assessment using bayesian updating'. Together they form a unique fingerprint.

Cite this