Saved in:
Bibliographic Details
Main Authors: Chen, Yinsong, Wang, Kaifeng, Meng, Xiaoqiang, Li, Xueyuan, Li, Zirui, Gao, Xin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.15587
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912494626275328
author Chen, Yinsong
Wang, Kaifeng
Meng, Xiaoqiang
Li, Xueyuan
Li, Zirui
Gao, Xin
author_facet Chen, Yinsong
Wang, Kaifeng
Meng, Xiaoqiang
Li, Xueyuan
Li, Zirui
Gao, Xin
contents Current research on decision-making in safety-critical scenarios often relies on inefficient data-driven scenario generation or specific modeling approaches, which fail to capture corner cases in real-world contexts. To address this issue, we propose a Red-Team Multi-Agent Reinforcement Learning framework, where background vehicles with interference capabilities are treated as red-team agents. Through active interference and exploration, red-team vehicles can uncover corner cases outside the data distribution. The framework uses a Constraint Graph Representation Markov Decision Process, ensuring that red-team vehicles comply with safety rules while continuously disrupting the autonomous vehicles (AVs). A policy threat zone model is constructed to quantify the threat posed by red-team vehicles to AVs, inducing more extreme actions to increase the danger level of the scenario. Experimental results show that the proposed framework significantly impacts AVs decision-making safety and generates various corner cases. This method also offers a novel direction for research in safety-critical scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2507_15587
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Red-Team Multi-Agent Reinforcement Learning for Emergency Braking Scenario
Chen, Yinsong
Wang, Kaifeng
Meng, Xiaoqiang
Li, Xueyuan
Li, Zirui
Gao, Xin
Machine Learning
Artificial Intelligence
Current research on decision-making in safety-critical scenarios often relies on inefficient data-driven scenario generation or specific modeling approaches, which fail to capture corner cases in real-world contexts. To address this issue, we propose a Red-Team Multi-Agent Reinforcement Learning framework, where background vehicles with interference capabilities are treated as red-team agents. Through active interference and exploration, red-team vehicles can uncover corner cases outside the data distribution. The framework uses a Constraint Graph Representation Markov Decision Process, ensuring that red-team vehicles comply with safety rules while continuously disrupting the autonomous vehicles (AVs). A policy threat zone model is constructed to quantify the threat posed by red-team vehicles to AVs, inducing more extreme actions to increase the danger level of the scenario. Experimental results show that the proposed framework significantly impacts AVs decision-making safety and generates various corner cases. This method also offers a novel direction for research in safety-critical scenarios.
title Red-Team Multi-Agent Reinforcement Learning for Emergency Braking Scenario
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2507.15587