Constraint-Aware Diffusion Guidance for Robotics: Real-Time Obstacle Avoidance for Autonomous Racing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ma, Hao, Bodmer, Sabrina, Carron, Andrea, Zeilinger, Melanie, Muehlebach, Michael
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912382150770688
author Ma, Hao
Bodmer, Sabrina
Carron, Andrea
Zeilinger, Melanie
Muehlebach, Michael
author_facet Ma, Hao
Bodmer, Sabrina
Carron, Andrea
Zeilinger, Melanie
Muehlebach, Michael
contents Diffusion models hold great potential in robotics due to their ability to capture complex, high-dimensional data distributions. However, their lack of constraint-awareness limits their deployment in safety-critical applications. We propose Constraint-Aware Diffusion Guidance (CoDiG), a data-efficient and general-purpose framework that integrates barrier functions into the denoising process, guiding diffusion sampling toward constraint-satisfying outputs. CoDiG enables constraint satisfaction even with limited training data and generalizes across tasks. We evaluate our framework in the challenging setting of miniature autonomous racing, where real-time obstacle avoidance is essential. Real-world experiments show that CoDiG generates safe outputs efficiently under dynamic conditions, highlighting its potential for broader robotic applications. A demonstration video is available at https://youtu.be/KNYsTdtdxOU.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13131
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Constraint-Aware Diffusion Guidance for Robotics: Real-Time Obstacle Avoidance for Autonomous Racing
Ma, Hao
Bodmer, Sabrina
Carron, Andrea
Zeilinger, Melanie
Muehlebach, Michael
Robotics
Systems and Control
Diffusion models hold great potential in robotics due to their ability to capture complex, high-dimensional data distributions. However, their lack of constraint-awareness limits their deployment in safety-critical applications. We propose Constraint-Aware Diffusion Guidance (CoDiG), a data-efficient and general-purpose framework that integrates barrier functions into the denoising process, guiding diffusion sampling toward constraint-satisfying outputs. CoDiG enables constraint satisfaction even with limited training data and generalizes across tasks. We evaluate our framework in the challenging setting of miniature autonomous racing, where real-time obstacle avoidance is essential. Real-world experiments show that CoDiG generates safe outputs efficiently under dynamic conditions, highlighting its potential for broader robotic applications. A demonstration video is available at https://youtu.be/KNYsTdtdxOU.
title Constraint-Aware Diffusion Guidance for Robotics: Real-Time Obstacle Avoidance for Autonomous Racing
topic Robotics
Systems and Control
url https://arxiv.org/abs/2505.13131