AdvDiffuser: Generating Adversarial Safety-Critical Driving Scenarios via Guided Diffusion

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xie, Yuting, Guo, Xianda, Wang, Cong, Liu, Kunhua, Chen, Long
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914969977618432
author Xie, Yuting
Guo, Xianda
Wang, Cong
Liu, Kunhua
Chen, Long
author_facet Xie, Yuting
Guo, Xianda
Wang, Cong
Liu, Kunhua
Chen, Long
contents Safety-critical scenarios are infrequent in natural driving environments but hold significant importance for the training and testing of autonomous driving systems. The prevailing approach involves generating safety-critical scenarios automatically in simulation by introducing adversarial adjustments to natural environments. These adjustments are often tailored to specific tested systems, thereby disregarding their transferability across different systems. In this paper, we propose AdvDiffuser, an adversarial framework for generating safety-critical driving scenarios through guided diffusion. By incorporating a diffusion model to capture plausible collective behaviors of background vehicles and a lightweight guide model to effectively handle adversarial scenarios, AdvDiffuser facilitates transferability. Experimental results on the nuScenes dataset demonstrate that AdvDiffuser, trained on offline driving logs, can be applied to various tested systems with minimal warm-up episode data and outperform other existing methods in terms of realism, diversity, and adversarial performance.
format Preprint
id arxiv_https___arxiv_org_abs_2410_08453
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AdvDiffuser: Generating Adversarial Safety-Critical Driving Scenarios via Guided Diffusion
Xie, Yuting
Guo, Xianda
Wang, Cong
Liu, Kunhua
Chen, Long
Machine Learning
Robotics
Safety-critical scenarios are infrequent in natural driving environments but hold significant importance for the training and testing of autonomous driving systems. The prevailing approach involves generating safety-critical scenarios automatically in simulation by introducing adversarial adjustments to natural environments. These adjustments are often tailored to specific tested systems, thereby disregarding their transferability across different systems. In this paper, we propose AdvDiffuser, an adversarial framework for generating safety-critical driving scenarios through guided diffusion. By incorporating a diffusion model to capture plausible collective behaviors of background vehicles and a lightweight guide model to effectively handle adversarial scenarios, AdvDiffuser facilitates transferability. Experimental results on the nuScenes dataset demonstrate that AdvDiffuser, trained on offline driving logs, can be applied to various tested systems with minimal warm-up episode data and outperform other existing methods in terms of realism, diversity, and adversarial performance.
title AdvDiffuser: Generating Adversarial Safety-Critical Driving Scenarios via Guided Diffusion
topic Machine Learning
Robotics
url https://arxiv.org/abs/2410.08453