arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.20179cs.HC
收藏

基于风险增强的结构化模式:面向LLM驱动的在线节点级决策

RESPOND: Risk-Enhanced Structured Pattern for LLM-driven Online Node-level Decision-making

Tsinghua University(清华大学);Massachusetts Institute of Technology(麻省理工学院);Singapore-MIT Alliance for Research and Technology (SMART)(新加坡-麻省理工学院研究与技术创新联盟);University of Wisconsin-Madison(威斯康星大学麦迪逊分校);Sun Yat-sen University(中山大学);China Mobile Group Guangdong Co., Ltd.(中国移动通信集团广东有限公司);Korea Advanced Institute of Science and Technology(韩国科学技术院)

Dan Chen, Heye Huang, Tiantian Chen, Zheng Li, Yongji Li, Yuhui Xu, Sikai Chen

AI总结:

RESPOND通过结构化模式和风险增强机制提升LLM驱动驾驶代理的决策精度与安全性,有效减少碰撞并实现个性化驾驶风格适应。

摘要:

当前依赖非结构化纯文本记忆的基于LLM的驾驶代理在场景检索精度低和反思效率低方面存在不足。为了解决这一限制,我们提出了RESPOND,一种基于显式风险模式的结构化决策框架,用于LLM驱动的代理。RESPOND使用统一的5乘3矩阵表示每个以自身为中心的场景,该矩阵编码空间拓扑和道路约束,使空间风险配置的检索保持一致和可靠。基于此表示,开发了一种混合规则和LLM决策流程,具有两级记忆机制。在高风险情境中,精确模式匹配使已验证的动作能够快速安全地重用,而在低风险情境中,子模式匹配支持个性化驾驶风格适应。此外,一种基于模式的反思机制从碰撞和接近碰撞帧中抽象战术修正,以更新结构化记忆,实现一次碰撞到一般化学习。广泛的实验表明了RESPOND的有效性。在highway-env中,RESPOND在性能上优于最先进的基于LLM和强化学习的驾驶代理,同时产生显著更少的碰撞。通过逐步的人类反馈,代理在大约20个决策步骤内获得了一种Sporty驾驶风格,通过子模式抽象。为了现实验证,RESPOND在HighD数据集中提取的53个高风险切入场景上进行了评估。对于每个事件,干预立即在切入前应用,RESPOND重新决定驾驶动作。与记录的人类行为相比,RESPOND在84.9%的场景中减少了后续风险,展示了其在现实驾驶条件下的实际可行性。这些结果突显了RESPOND在自动驾驶、个性化驾驶辅助和主动危险缓解方面的潜力。

英文摘要:

Current LLM-based driving agents that rely on unstructured plain-text memory suffer from low-precision scene retrieval and inefficient reflection. To address this limitation, we present RESPOND, a structured decision-making framework for LLM-driven agents grounded in explicit risk patterns. RESPOND represents each ego-centric scene using a unified 5 by 3 matrix that encodes spatial topology and road constraints, enabling consistent and reliable retrieval of spatial risk configurations. Based on this representation, a hybrid rule and LLM decision pipeline is developed with a two-tier memory mechanism. In high-risk contexts, exact pattern matching enables rapid and safe reuse of verified actions, while in low-risk contexts, sub-pattern matching supports personalized driving style adaptation. In addition, a pattern-aware reflection mechanism abstracts tactical corrections from crash and near-miss frames to update structured memory, achieving one-crash-to-generalize learning. Extensive experiments demonstrate the effectiveness of RESPOND. In highway-env, RESPOND outperforms state-of-the-art LLM-based and reinforcement learning based driving agents while producing substantially fewer collisions. With step-wise human feedback, the agent acquires a Sporty driving style within approximately 20 decision steps through sub-pattern abstraction. For real-world validation, RESPOND is evaluated on 53 high-risk cut-in scenarios extracted from the HighD dataset. For each event, intervention is applied immediately before the cut-in and RESPOND re-decides the driving action. Compared to recorded human behavior, RESPOND reduces subsequent risk in 84.9 percent of scenarios, demonstrating its practical feasibility under real-world driving conditions. These results highlight RESPONDs potential for autonomous driving, personalized driving assistance, and proactive hazard mitigation.

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。

↑