Bayesian Spatio-Temporal grAph tRansformer network (B-STAR) for multi-aircraft trajectory prediction

作者：

Highlights：

• We propose an uncertainty-aware multi-agent trajectory prediction model, B-STAR.

• B-STAR achieves state-of-the-art prediction performance on the ETH/UCY pedestrian dataset.

• B-STAR is adapted for near-terminal multi-aircraft trajectory prediction with ASDE-X radar recording data.

• We propose the machine learning problem-solving pipeline with advanced software tools.

• We show the effectiveness of the proposed framework with visualizations.

摘要

•We propose an uncertainty-aware multi-agent trajectory prediction model, B-STAR.•B-STAR achieves state-of-the-art prediction performance on the ETH/UCY pedestrian dataset.•B-STAR is adapted for near-terminal multi-aircraft trajectory prediction with ASDE-X radar recording data.•We propose the machine learning problem-solving pipeline with advanced software tools.•We show the effectiveness of the proposed framework with visualizations.

论文关键词：Multi-agent trajectory prediction,Graph neural network,Transformer,Air traffic management

论文评审过程：Received 16 December 2021, Revised 3 May 2022, Accepted 4 May 2022, Available online 11 May 2022, Version of Record 20 May 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.108998