LAformer

Trajectory Prediction for Autonomous Driving with Lane-Aware Scene Constraints

authored by

Mengmeng Liu, Hao Cheng, Lin Chen, Hellward Broszio, Jiangtao Li, Runjiang Zhao, Monika Sester, Michael Ying Yang

Abstract

Existing trajectory prediction methods for autonomous driving typically rely on one-stage trajectory prediction models, which condition future trajectories on observed trajectories combined with fused scene information. However, they often struggle with complex scene constraints, such as those encountered at intersections. To this end, we present a novel method, called LAformer. It uses an attention-based temporally dense lane-aware estimation module to continuously estimate the likelihood of the alignment between motion dynamics and scene information extracted from an HD map. Additionally, unlike one-stage prediction models, LAformer utilizes predictions from the first stage as anchor trajectories. It leverages a second-stage motion refinement module to further explore temporal consistency across the complete time horizon. Extensive experiments on nuScenes and Argoverse 1 demonstrate that LAformer achieves excellent generalized performance for multimodal trajectory prediction. The source code of LAformer is available at github.com/mengmengliu1998/LAformer.

Organisation(s)

Institute of Cartography and Geoinformatics
Graduiertenkolleg 2159: Integrität und Kollaboration in dynamischen Sensornetzen

External Organisation(s)

University of Twente
VISCODA GmbH
PhiGent Robotics
University of Bath

Type

Conference contribution

Pages

2039-2049

No. of pages

Publication date

17.06.2024

Publication status

Published

Peer reviewed

Yes

ASJC Scopus subject areas

Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

Electronic version(s)

https://doi.org/10.48550/arXiv.2302.13933 (Access: Open)
https://doi.org/10.1109/CVPRW63382.2024.00209 (Access: Closed)

BibTeX

@inproceedings{9631cb9c26594b4d88bd1d86fae3400c,
title = "LAformer: Trajectory Prediction for Autonomous Driving with Lane-Aware Scene Constraints",
abstract = "Existing trajectory prediction methods for autonomous driving typically rely on one-stage trajectory prediction models, which condition future trajectories on observed trajectories combined with fused scene information. However, they often struggle with complex scene constraints, such as those encountered at intersections. To this end, we present a novel method, called LAformer. It uses an attention-based temporally dense lane-aware estimation module to continuously estimate the likelihood of the alignment between motion dynamics and scene information extracted from an HD map. Additionally, unlike one-stage prediction models, LAformer utilizes predictions from the first stage as anchor trajectories. It leverages a second-stage motion refinement module to further explore temporal consistency across the complete time horizon. Extensive experiments on nuScenes and Argoverse 1 demonstrate that LAformer achieves excellent generalized performance for multimodal trajectory prediction. The source code of LAformer is available at https://github.com/mengmengliu1998/LAformer.",
keywords = "lane-aware selection, motion refinement, multimodal, Trajectory prediction",
author = "Mengmeng Liu and Hao Cheng and Lin Chen and Hellward Broszio and Jiangtao Li and Runjiang Zhao and Monika Sester and Yang, {Michael Ying}",
note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2024 ; Conference date: 16-06-2024 Through 22-06-2024",
year = "2024",
month = jun,
day = "17",
doi = "10.48550/arXiv.2302.13933",
language = "English",
isbn = "979-8-3503-6548-1",
series = "IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",
publisher = "IEEE Computer Society",
pages = "2039--2049",
booktitle = "2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)",
address = "United States",
}