Multi-Head Attention based Probabilistic Vehicle Trajectory Prediction

This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple encoder-decoder architecture based on multihead attention. The proposed model generates the distribution of the predicted trajectories for multiple vehicles in parallel. Our ap...

Full description

Saved in:
Bibliographic Details
Main Authors: Kim, Hayoung, Kim, Dongchan, Kim, Gihoon, Cho, Jeongmin, Huh, Kunsoo
Format: Conference Proceeding
Language:English
Subjects:
Decoding
Predictive models
Probabilistic logic
Roads
Training
Trajectory
Uncertainty
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple encoder-decoder architecture based on multihead attention. The proposed model generates the distribution of the predicted trajectories for multiple vehicles in parallel. Our approach to model the interactions can learn to attend to a few influential vehicles in an unsupervised manner, which can improve the interpretability of the network. The experiments using naturalistic trajectories at highway show the clear improvement in terms of positional error on both longitudinal and lateral direction.
ISSN:2642-7214
DOI:10.1109/IV47402.2020.9304741