Probabilistic aircraft trajectory prediction with weather uncertainties using approximate bayesian variational inference to neural networks

Yutian Pang, Yuhao Wang, Yongming Liu

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

Abstract

A key consideration in Trajectory Prediction (TP) tools is the confidence that can be placed on the prediction. We purpose a non-deterministic TP neural network using tractable approximate Bayesian variational inference for the model parameters considering weather effects. This work adopts the state-of-art in Bayesian Deep Learning research and builds a neural network model with stochastic convolutional, recurrent, and fully-connect layers. The purposed stochastic variational method outperforms the dropout approximate to Variational Inference and performs reliable uncertainty estimates. It can be easily applied to most neural net architectures and also provides a simple pruning heuristic that can drastically reduce the number of model parameters compares to ensemble methods. The experiment is conducted with the Atlanta Air Route Traffic Control Center (ZTL) flight data and the corridor integrated weather system (CIWS) weather data from Sherlock Data Warehouse (SDW) on June 24th, 2019. The experimental results show better variance reduction than dropout-based methods. The uncertainty estimates are more reliable thanks to the Kullback–Leibler divergence (KL-divergence) term within the optimization objective.

Original languageEnglish (US)
Title of host publicationAIAA AVIATION 2020 FORUM
PublisherAmerican Institute of Aeronautics and Astronautics Inc, AIAA
ISBN (Print)9781624105982
DOIs
StatePublished - 2020
EventAIAA AVIATION 2020 FORUM - Virtual, Online
Duration: Jun 15 2020Jun 19 2020

Publication series

NameAIAA AVIATION 2020 FORUM
Volume1 PartF

Conference

ConferenceAIAA AVIATION 2020 FORUM
CityVirtual, Online
Period6/15/206/19/20

ASJC Scopus subject areas

  • Nuclear Energy and Engineering
  • Aerospace Engineering
  • Energy Engineering and Power Technology

Fingerprint Dive into the research topics of 'Probabilistic aircraft trajectory prediction with weather uncertainties using approximate bayesian variational inference to neural networks'. Together they form a unique fingerprint.

Cite this