基于LLM常识推理的多智能体编排自动驾驶框架

AI 摘要

该研究提出一种混合自动驾驶框架，通过编排器协调PPO强化学习与PID控制，并利用大语言模型（LLM）的常识推理能力作为顾问和奖励机制，而非直接控制车辆。在CARLA模拟器的随机场景中评估，结果显示该方法在保持实时性和安全性的同时，提升了上下文推理能力。

正文节选

Multi-Agent Orchestration with the Common-Sense Reasoning Capabilities of LLMs for Autonomous Driving Abstract Autonomous vehicles require robust perception and decision-making capabilities to operate in diverse and unseen scenarios. While reinforcement learning and rule-based methods can provide effective control and safety mechanisms, their performance may degrade in situations requiring contextual reasoning. Large Language Models (LLMs) have demonstrated strong capabilities in understanding m

发布时间：2026-08-21 12:00

抓取时间：2026-08-21 12:03

来源机构：arXiv