发表机构
Seoul National University（首尔大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有驾驶基准短视与导航指令模糊的问题，提出Odyssey闭环基准，用显式SD路线提供导航目标，并引入新指标评估长时域驾驶中的路线遵循与变道准备，揭示端到端驾驶的开放问题。
AI 中文摘要
端到端驾驶的闭环评估需要连续的滚动仿真，以揭示早期决策如何影响后续驾驶行为。然而，现有基准仅评估短片段，无法捕捉后续后果。模糊的方向指令也掩盖了预期的导航目标。我们引入了Odyssey，一个用于长时域驾驶的闭环基准，包含100个场景，每个场景均从100秒的nuPlan驾驶日志重建，以保留导航操作和交通交互的上下文。为提供一致的导航目标，Odyssey将方向指令替换为显式的标准定义（SD）地图路线，指定应遵循的道路，而基于传感器的规划则决定局部驾驶动作。在这些滚动仿真中，基于扩散的3DGS渲染图像细化减少了沿自车轨迹的渲染伪影。为评估规划器遵循这些路线并为即将到来的操作做准备的效果，我们引入了SD路线合规性和变道前得分。这些评估辅以RouteDS，它扩展了驾驶得分，增加了对SD路线偏离和变道准备失败的惩罚。我们改编了最先进的规划器，包括视觉-语言-动作（VLA）模型，并使用这些指标评估其导航性能。Odyssey突显了端到端驾驶中路线表示与集成的开放问题。基准代码和改编的基线将公开发布。
英文摘要
Closed-loop evaluation of end-to-end driving requires continuous rollouts that reveal how earlier decisions affect subsequent driving. However, existing benchmarks evaluate only short segments and fail to capture later consequences. Ambiguous directional commands also obscure the intended navigation objective. We introduce Odyssey, a closed-loop benchmark for long-horizon driving comprising 100 scenarios, each reconstructed from a 100-second nuPlan driving log to preserve the context of navigation maneuvers and traffic interactions. To provide a consistent navigation objective, Odyssey replaces directional commands with explicit standard-definition (SD) map routes that specify which roads to follow, while sensor-based planning determines local driving actions. Throughout these rollouts, diffusion-based refinement of 3DGS-rendered images reduces rendering artifacts along the ego trajectory. To assess how effectively planners follow these routes and prepare for upcoming maneuvers, we introduce SD Route Compliance and Pre-Lane Change Score. These assessments are complemented by RouteDS, which extends the Driving Score with penalties for SD-route deviations and failed lane preparation. We adapt state-of-the-art planners, including vision-language-action (VLA) models, and evaluate their navigation performance using these metrics. Odyssey highlights open questions in route representation and integration for E2E driving. Benchmark code and adapted baselines will be released publicly.
Comments26pages, 12 figures
