arXiv · cs.AI · Long Zhang, Yuchen Xia, Bingqing Wei, Zhen Liu, Shiwen Mao, Zhu Han, Mohsen Guizani · 5 小时前模型评分32
大模型如何驱动具身智能驾驶：语义与策略双驱动混合决策框架
Large Multimodal Models for Embodied Intelligent Driving: Enabling the Next Generation of Autonomous Mobility
导读
一项研究提出语义与策略双驱动的混合决策框架，将大语言模型与多模态模型（LMM，Large Multimodal Model）的语义理解能力与深度强化学习的实时策略优化结合，用于具身智能驾驶。在换道规划案例的仿真中，该框架比基线方法至少提升 19% 的性能。研究同时梳理了该方向的开放挑战与未来研究方向。
分类置信度
中 · 待人工复核
- 存在未解决的相邻类别：embodied-ai，需要人工复核。
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
应用领域：交通与物流
事件类型：研究发布
证据摘录
- “This article proposes a novel semantics and policy dual-driven hybrid decision framework to tackle this challenge, ensuring continuous learning and joint decision.”
- “The proposed framework, evaluated through a lane-change planning case study, achieves at least a 19% performance improvement over the baselines, as verified by simulations.”
来源：arXiv · cs.AI · arxiv.org