SafeCoop: Unravelling Full Stack Safety in Agentic Collaborative Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Xiangbo, Lin, Tzu-Hsiang, Song, Ruojing, Wu, Yuheng, Huang, Kuan-Ru, Jin, Zicheng, Lin, Fangzhou, Liu, Shinan, Tu, Zhengzhong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917029289656320
author Gao, Xiangbo
Lin, Tzu-Hsiang
Song, Ruojing
Wu, Yuheng
Huang, Kuan-Ru
Jin, Zicheng
Lin, Fangzhou
Liu, Shinan
Tu, Zhengzhong
author_facet Gao, Xiangbo
Lin, Tzu-Hsiang
Song, Ruojing
Wu, Yuheng
Huang, Kuan-Ru
Jin, Zicheng
Lin, Fangzhou
Liu, Shinan
Tu, Zhengzhong
contents Collaborative driving systems leverage vehicle-to-everything (V2X) communication across multiple agents to enhance driving safety and efficiency. Traditional V2X systems take raw sensor data, neural features, or perception results as communication media, which face persistent challenges, including high bandwidth demands, semantic loss, and interoperability issues. Recent advances investigate natural language as a promising medium, which can provide semantic richness, decision-level reasoning, and human-machine interoperability at significantly lower bandwidth. Despite great promise, this paradigm shift also introduces new vulnerabilities within language communication, including message loss, hallucinations, semantic manipulation, and adversarial attacks. In this work, we present the first systematic study of full-stack safety and security issues in natural-language-based collaborative driving. Specifically, we develop a comprehensive taxonomy of attack strategies, including connection disruption, relay/replay interference, content spoofing, and multi-connection forgery. To mitigate these risks, we introduce an agentic defense pipeline, which we call SafeCoop, that integrates a semantic firewall, language-perception consistency checks, and multi-source consensus, enabled by an agentic transformation function for cross-frame spatial alignment. We systematically evaluate SafeCoop in closed-loop CARLA simulation across 32 critical scenarios, achieving 69.15% driving score improvement under malicious attacks and up to 67.32% F1 score for malicious detection. This study provides guidance for advancing research on safe, secure, and trustworthy language-driven collaboration in transportation systems. Our project page is https://xiangbogaobarry.github.io/SafeCoop.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18123
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SafeCoop: Unravelling Full Stack Safety in Agentic Collaborative Driving
Gao, Xiangbo
Lin, Tzu-Hsiang
Song, Ruojing
Wu, Yuheng
Huang, Kuan-Ru
Jin, Zicheng
Lin, Fangzhou
Liu, Shinan
Tu, Zhengzhong
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Robotics
Collaborative driving systems leverage vehicle-to-everything (V2X) communication across multiple agents to enhance driving safety and efficiency. Traditional V2X systems take raw sensor data, neural features, or perception results as communication media, which face persistent challenges, including high bandwidth demands, semantic loss, and interoperability issues. Recent advances investigate natural language as a promising medium, which can provide semantic richness, decision-level reasoning, and human-machine interoperability at significantly lower bandwidth. Despite great promise, this paradigm shift also introduces new vulnerabilities within language communication, including message loss, hallucinations, semantic manipulation, and adversarial attacks. In this work, we present the first systematic study of full-stack safety and security issues in natural-language-based collaborative driving. Specifically, we develop a comprehensive taxonomy of attack strategies, including connection disruption, relay/replay interference, content spoofing, and multi-connection forgery. To mitigate these risks, we introduce an agentic defense pipeline, which we call SafeCoop, that integrates a semantic firewall, language-perception consistency checks, and multi-source consensus, enabled by an agentic transformation function for cross-frame spatial alignment. We systematically evaluate SafeCoop in closed-loop CARLA simulation across 32 critical scenarios, achieving 69.15% driving score improvement under malicious attacks and up to 67.32% F1 score for malicious detection. This study provides guidance for advancing research on safe, secure, and trustworthy language-driven collaboration in transportation systems. Our project page is https://xiangbogaobarry.github.io/SafeCoop.
title SafeCoop: Unravelling Full Stack Safety in Agentic Collaborative Driving
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Robotics
url https://arxiv.org/abs/2510.18123