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 2603.07901cs.ROcs.LG
收藏

NaviDriveVLM:解耦高层推理与运动规划用于自动驾驶

NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving

J. Mike Walker ’66 Department of Mechanical Engineering, Texas A&M University, College Station, TX 77843, USA(杰米·沃克’66机械工程系,德克萨斯A&M大学,学院站,TX 77843,美国);The Department of Engineering Technology and Industrial Distribution Texas A&M University, College Station, TX 77843, USA(工程技术与工业分配系,德克萨斯A&M大学,学院站,TX 77843,美国)

Ximeng Tao, Pardis Taghavi, Dimitar Filev, Reza Langari, Gaurav Pandey

AI总结:

NaviDriveVLM通过解耦高层推理与运动规划,提升自动驾驶中的推理能力与效率

摘要:

视觉-语言模型(VLMs)通过联合建模视觉观测、驾驶上下文和基于语言的推理,成为端到端自动驾驶(AD)的有前途方向。然而,现有基于VLM的系统面临高层推理与运动规划之间的权衡:大模型提供强大的语义理解,但适应精确控制成本高;小VLM模型可高效微调,但推理能力较弱。我们提出NaviDriveVLM,一种解耦框架,通过大规模导航器和轻量级可训练驾驶器分离推理与动作生成。该设计保留推理能力,降低训练成本,并为下游规划提供显式的可解释中间表示。在nuScenes基准测试中,NaviDriveVLM在端到端运动规划中优于大VLM基线。

英文摘要:

Vision-language models (VLMs) have emerged as a promising direction for end-to-end autonomous driving (AD) by jointly modeling visual observations, driving context, and language-based reasoning. However, existing VLM-based systems face a trade-off between high-level reasoning and motion planning: large models offer strong semantic understanding but are costly to adapt for precise control, whereas small VLM models can be fine-tuned efficiently but often exhibit weaker reasoning. We propose NaviDriveVLM, a decoupled framework that separates reasoning from action generation using a large-scale Navigator and a lightweight trainable Driver. This design preserves reasoning ability, reduces training cost, and provides an explicit interpretable intermediate representation for downstream planning. Experiments on the nuScenes benchmark show that NaviDriveVLM outperforms large VLM baselines in end-to-end motion planning.

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。

↑