GuideFlow：面向端到端自动驾驶规划的约束引导流匹配
GuideFlow: Constraint-Guided Flow Matching for Planning in End-to-End Autonomous Driving
把流匹配生成过程与能量模型联合训练，在采样阶段直接施加显式物理/安全约束，可作为视觉规划器中替代扩散策略、避免轨迹模式坍缩并省去后处理优化阶段的通用生成式规划范式。
原文依据 GuideFlow unifies the training of the flow matching with the Energy-Based Model (EBM)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论驾驶规划是端到端（E2E）自动驾驶的关键组成部分。然而，主流的模仿式端到端规划器常常受困于多模态轨迹的模式坍缩，无法生成多样化的轨迹候选。与此同时，生成式端到端规划器难以将关键的安全与物理约束直接纳入生成过程，因而需要额外的优化阶段来精修其输出。本文提出 GuideFlow，一种利用约束流匹配的新型规划框架。具体而言，GuideFlow 显式地对流匹配过程建模，该过程本身即可缓解模式坍缩，并允许来自各类条件信号的灵活引导。我们的核心贡献在于直接在流匹配生成过程中施加显式约束，而非依赖隐式的约束编码。关键在于，GuideFlow 将流匹配的训练与能量模型（EBM）统一起来，以增强模型自主优化的能力，从而稳健地满足物理约束。其次，GuideFlow 将驾驶激进度参数化为生成过程中的控制信号，从而能够精确操控轨迹风格。在主要驾驶基准（Bench2Drive、NuScenes、NavSim 与 ADV-NuScenes）上的大量评估验证了 GuideFlow 的有效性。值得注意的是，在 NavSim 测试困难划分（Navhard）上，GuideFlow 以 43.0 的 EPDMS 得分取得了 SOTA。代码将发布于 https://github.com/adept-thu/GuideFlow。
英文原文摘要 1542 字符
Driving planning is a critical component of end-to-end (E2E) autonomous driving. However, prevailing Imitative E2E Planners often suffer from multimodal trajectory mode collapse, failing to produce diverse trajectory proposals. Meanwhile, Generative E2E Planners struggle to incorporate crucial safety and physical constraints directly into the generative process, necessitating an additional optimization stage to refine their outputs. In this paper, we propose GuideFlow, a novel planning framework that leverages Constrained Flow Matching. Concretely, GuideFlow explicitly models the flow matching process, which inherently mitigates mode collapse and allows for flexible guidance from various conditioning signals. Our core contribution lies in directly enforcing explicit constraints within the flow matching generation process, rather than relying on implicit constraint encoding. Crucially, GuideFlow unifies the training of the flow matching with the Energy-Based Model (EBM) to enhance the model's autonomous optimization capability to robustly satisfy physical constraints. Secondly, GuideFlow parameterizes driving aggressiveness as a control signal during generation, enabling precise manipulation of trajectory style. Extensive evaluations on major driving benchmarks (Bench2Drive, NuScenes, NavSim and ADV-NuScenes) validate the effectiveness of GuideFlow. Notably, on the NavSim test hard split (Navhard), GuideFlow achieved SOTA with an EPDMS score of 43.0. The code will be released in https://github.com/adept-thu/GuideFlow.