用大模型协同控制匝道合流，提升复杂路况下的驾驶安全与效率。
A Cascading Cooperative Multi-agent Framework for On-ramp Merging Control Integrating Large Language Models
- 分层协作框架：强化学习处理个体行为，大模型负责区域协调。
- 相比传统方法，微观与宏观性能均显著提升，合流成功率更高。
- 适合自动驾驶、智能交通系统研究者，尤其关注多车协同决策。
传统强化学习在复现人类驾驶行为、多智能体场景泛化及可解释性方面存在挑战，尤其在需要深度环境理解、智能体协同与动态优化的复杂场景中更为突出。尽管大语言模型（LLM）在泛化与互操作性方面表现良好，但常忽略必要的多智能体协同。为此，我们提出分层协同多智能体（CCMA）框架，融合强化学习用于个体交互，微调后的大型语言模型实现区域协作，设计奖励函数完成全局优化，并引入检索增强生成机制，在复杂驾驶场景中动态优化决策。实验表明，该框架在复杂驾驶环境中显著优于现有强化学习方法，微观与宏观性能均有明显提升。
原文摘要 · Abstract (English)
Traditional Reinforcement Learning (RL) suffers from replicating human-like behaviors, generalizing effectively in multi-agent scenarios, and overcoming inherent interpretability issues.These tasks are compounded when deep environment understanding, agent coordination and dynamic optimization are required. While Large Language Model (LLM) enhanced methods have shown promise in generalization and interoperability, they often neglect necessary multi-agent coordination. Therefore, we introduce the Cascading Cooperative Multi-agent (CCMA) framework, integrating RL for individual interactions, a fine-tuned LLM for regional cooperation, a reward function for global optimization, and the Retrieval-augmented Generation mechanism to dynamically optimize decision-making across complex driving scenarios. Our experiments demonstrate that the CCMA outperforms existing RL methods, demonstrating significant improvements in both micro and macro-level performance in complex driving environments.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。