• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
PRESTO: Fast Motion Planning Using Diffusion Models Based on Key-Configuration Environment Representation.
Mingyo
Seo, Yoonyoung Cho, Yoonchang Sung, Peter
Stone, Yuke Zhu, and Beomjoon Kim.
In IEEE International Conference
on Robotics and Automation (ICRA), May 2025.
[PDF]1.4MB [slides.pdf]2.1MB [poster.pdf]1.3MB
We introduce a learning-guided motion planning framework that generates seedtrajectories using a diffusion model for trajectory optimization. Given aworkspace, our method approximates the configuration space (C-space) ob- staclesthrough an environment representation consisting of a sparse set of task-relatedkey configurations, which is then used as a conditioning input to the diffusionmodel. The diffusion model integrates regularization terms that encourage smooth,collision-free trajectories during training, and trajectory op- timizationrefines the generated seed trajectories to correct any colliding segments. Ourexperimental results demonstrate that high-quality trajectory priors, learnedthrough our C- space-grounded diffusion model, enable the efficient generation ofcollision-free trajectories in narrow-passage environments, outperformingprevious learning- and planning-based baselines. Videos and additional materialscan be found on the project page: https://kiwi-sherbet.github.io/PRESTO.
@InProceedings{mingyo_seo_ICRA2025,
author = {Mingyo Seo and Yoonyoung Cho and Yoonchang Sung and Peter Stone and Yuke Zhu and Beomjoon Kim},
title = {PRESTO: Fast Motion Planning Using Diffusion Models Based on Key-Configuration Environment Representation},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
year = {2025},
month = {May},
location = {Atlanta, United States},
abstract = {We introduce a learning-guided motion planning framework that generates seed
trajectories using a diffusion model for trajectory optimization. Given a
workspace, our method approximates the configuration space (C-space) ob- stacles
through an environment representation consisting of a sparse set of task-related
key configurations, which is then used as a conditioning input to the diffusion
model. The diffusion model integrates regularization terms that encourage smooth,
collision-free trajectories during training, and trajectory op- timization
refines the generated seed trajectories to correct any colliding segments. Our
experimental results demonstrate that high-quality trajectory priors, learned
through our C- space-grounded diffusion model, enable the efficient generation of
collision-free trajectories in narrow-passage environments, outperforming
previous learning- and planning-based baselines. Videos and additional materials
can be found on the project page: https://kiwi-sherbet.github.io/PRESTO.
},
}
Generated by bib2html.pl (written by Patrick Riley ) on Sat Nov 01, 2025 23:24:55