TrajDiffuse: A Conditional Diffusion Model for Environment-Aware Trajectory Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qingze, Liu, Li, Danrui, Sohn, Samuel S., Yoon, Sejong, Kapadia, Mubbasir, Pavlovic, Vladimir
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915141873827840
author Qingze
Liu
Li, Danrui
Sohn, Samuel S.
Yoon, Sejong
Kapadia, Mubbasir
Pavlovic, Vladimir
author_facet Qingze
Liu
Li, Danrui
Sohn, Samuel S.
Yoon, Sejong
Kapadia, Mubbasir
Pavlovic, Vladimir
contents Accurate prediction of human or vehicle trajectories with good diversity that captures their stochastic nature is an essential task for many applications. However, many trajectory prediction models produce unreasonable trajectory samples that focus on improving diversity or accuracy while neglecting other key requirements, such as collision avoidance with the surrounding environment. In this work, we propose TrajDiffuse, a planning-based trajectory prediction method using a novel guided conditional diffusion model. We form the trajectory prediction problem as a denoising impaint task and design a map-based guidance term for the diffusion process. TrajDiffuse is able to generate trajectory predictions that match or exceed the accuracy and diversity of the SOTA, while adhering almost perfectly to environmental constraints. We demonstrate the utility of our model through experiments on the nuScenes and PFSD datasets and provide an extensive benchmark analysis against the SOTA methods.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10804
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TrajDiffuse: A Conditional Diffusion Model for Environment-Aware Trajectory Prediction
Qingze
Liu
Li, Danrui
Sohn, Samuel S.
Yoon, Sejong
Kapadia, Mubbasir
Pavlovic, Vladimir
Computer Vision and Pattern Recognition
Machine Learning
Accurate prediction of human or vehicle trajectories with good diversity that captures their stochastic nature is an essential task for many applications. However, many trajectory prediction models produce unreasonable trajectory samples that focus on improving diversity or accuracy while neglecting other key requirements, such as collision avoidance with the surrounding environment. In this work, we propose TrajDiffuse, a planning-based trajectory prediction method using a novel guided conditional diffusion model. We form the trajectory prediction problem as a denoising impaint task and design a map-based guidance term for the diffusion process. TrajDiffuse is able to generate trajectory predictions that match or exceed the accuracy and diversity of the SOTA, while adhering almost perfectly to environmental constraints. We demonstrate the utility of our model through experiments on the nuScenes and PFSD datasets and provide an extensive benchmark analysis against the SOTA methods.
title TrajDiffuse: A Conditional Diffusion Model for Environment-Aware Trajectory Prediction
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2410.10804