SafeCoop：揭示智能体协同驾驶中的全栈安全问题
SafeCoop: Unravelling Full Stack Safety in Agentic Collaborative Driving
AI总结：
该研究首次系统探究基于自然语言的协同驾驶全栈安全问题，提出集成语义防火墙等模块的SafeCoop智能体防御方案，在CARLA仿真32个场景中显著提升攻击下的驾驶得分与恶意检测效果。
摘要：
协同驾驶系统利用多智能体间的车联网（V2X）通信来提升驾驶安全性与效率。传统V2X系统以原始传感器数据、神经特征或感知结果作为通信介质，长期面临高带宽需求、语义损失、互操作性差等挑战。近期研究探索将自然语言作为一种极具潜力的通信介质，它能以显著更低的带宽提供丰富语义、决策级推理能力与人机互操作性。尽管前景广阔，这一范式转变也在语言通信中引入了新的安全漏洞，包括消息丢失、幻觉、语义篡改与对抗性攻击。\n本研究首次针对基于自然语言的协同驾驶中的全栈安全与安保问题开展系统性研究。具体而言，我们构建了一套全面的攻击策略分类体系，涵盖连接中断、中继/重放干扰、内容欺骗、多连接伪造四类。为缓解这些风险，我们提出了一套名为SafeCoop的智能体防御流水线，它集成了语义防火墙、语言-感知一致性检查与多源共识机制，并通过智能体转换函数实现跨帧空间对齐。我们在闭环CARLA仿真环境的32个关键场景中对SafeCoop进行了系统性评估，结果显示其在恶意攻击下驾驶得分提升69.15%，恶意检测的F1分数最高可达67.32%。本研究为推进交通系统中安全、可靠、可信的语言驱动协同研究提供了指导，项目主页为https://xiangbogaobarry.github.io/SafeCoop。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
