SGDrive：面向自动驾驶的场景到目标分层世界认知
SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving
把通用 VLM 的表征按 scene-agent-goal 层级显式结构化，为视觉研究者提供了一种将通用视觉语言特征适配到具身决策任务的表征组织范式，可直接借鉴到需要时空推理的规划类任务。
原文依据 SGDrive decomposes driving understanding into a scene-agent-goal hierarchy that mirrors human driving cognition
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期端到端自动驾驶方法借助视觉语言模型（VLM）提升复杂驾驶场景下的规划能力。然而，VLM 本质上作为通用模型训练，缺乏对三维空间与时间中驾驶特定推理的专门理解。将其应用于自动驾驶时，这些模型难以建立结构化的时空表征，以刻画安全轨迹规划所必需的几何关系、场景上下文与运动模式。为解决上述局限，我们提出 SGDrive，一种围绕驾驶特定知识层级显式组织 VLM 表征学习的新框架。SGDrive 以预训练 VLM 为骨干，将驾驶理解分解为场景—智能体—目标层级，以对应人类驾驶认知：驾驶员首先感知整体环境（场景上下文），随后关注安全关键智能体及其行为，最后在执行动作前形成短期目标。该分层分解提供了通用 VLM 所缺乏的结构化时空表征，将多层级信息整合为紧凑而全面的形式用于轨迹规划。在 NAVSIM 基准上的大量实验表明，SGDrive 在纯视觉方法中取得最优性能，验证了分层知识结构化对将通用 VLM 适配到自动驾驶的有效性。
英文原文摘要 1512 字符
Recent end-to-end autonomous driving approaches have leveraged Vision-Language Models (VLMs) to enhance planning capabilities in complex driving scenarios. However, VLMs are inherently trained as generalist models, lacking specialized understanding of driving-specific reasoning in 3D space and time. When applied to autonomous driving, these models struggle to establish structured spatial-temporal representations that capture geometric relationships, scene context, and motion patterns critical for safe trajectory planning.To address these limitations, we propose SGDrive, a novel framework that explicitly structures the VLM's representation learning around driving-specific knowledge hierarchies. Built upon a pre-trained VLM backbone, SGDrive decomposes driving understanding into a scene-agent-goal hierarchy that mirrors human driving cognition: drivers first perceive the overall environment (scene context), then attend to safety-critical agents and their behaviors, and finally formulate short-term goals before executing actions. This hierarchical decomposition provides the structured spatial-temporal representation that generalist VLMs lack, integrating multi-level information into a compact yet comprehensive format for trajectory planning.Extensive experiments on the NAVSIM benchmark demonstrate that SGDrive achieves state-of-the-art performance among camera-only methods, validating the effectiveness of hierarchical knowledge structuring for adapting generalist VLMs to autonomous driving.