标题： SAFe-Copilot：统一共享自主框架 显示英文标题

标题： SAFe-Copilot: Unified Shared Autonomy Framework

摘要： 自动驾驶系统在罕见、模糊和分布外的场景中仍然脆弱，而在这些情况下人类驾驶员通过情境推理取得成功。 共享自主性作为一种有前途的方法出现，通过在自主性不确定时结合人类输入来减轻此类失败。 然而，大多数现有方法将仲裁限制在低级轨迹上，这些轨迹仅表示几何路径，因此无法保留底层的驾驶意图。 我们提出了一种统一的共享自主框架，在更高层次的抽象上整合人类输入和自主规划器。 我们的方法利用视觉语言模型（VLMs）从多模态线索——如驾驶员动作和环境上下文——推断驾驶员意图，并合成在人类和自主控制之间起中介作用的一致策略。 我们首先在模拟人类的环境中研究该框架，实现了完美的召回率，同时具有高准确率和精确度。 一项针对人类受试者的调查进一步表明高度一致，参与者在92%的情况下同意仲裁结果。 最后，在Bench2Drive基准上的评估表明，与纯自主相比，碰撞率显著降低，整体性能得到改善。 在语义的、基于语言的表示层面进行仲裁成为共享自主的设计原则，使系统能够进行常识推理并保持与人类意图的连续性。 显示英文摘要

摘要： Autonomous driving systems remain brittle in rare, ambiguous, and out-of-distribution scenarios, where human driver succeed through contextual reasoning. Shared autonomy has emerged as a promising approach to mitigate such failures by incorporating human input when autonomy is uncertain. However, most existing methods restrict arbitration to low-level trajectories, which represent only geometric paths and therefore fail to preserve the underlying driving intent. We propose a unified shared autonomy framework that integrates human input and autonomous planners at a higher level of abstraction. Our method leverages Vision Language Models (VLMs) to infer driver intent from multi-modal cues -- such as driver actions and environmental context -- and to synthesize coherent strategies that mediate between human and autonomous control. We first study the framework in a mock-human setting, where it achieves perfect recall alongside high accuracy and precision. A human-subject survey further shows strong alignment, with participants agreeing with arbitration outcomes in 92% of cases. Finally, evaluation on the Bench2Drive benchmark demonstrates a substantial reduction in collision rate and improvement in overall performance compared to pure autonomy. Arbitration at the level of semantic, language-based representations emerges as a design principle for shared autonomy, enabling systems to exercise common-sense reasoning and maintain continuity with human intent.