MeanFuser：基于MeanFlow的快速单步多模态轨迹生成与自适应重建用于端到端自动驾驶
MeanFuser: Fast One-Step Multi-Modal Trajectory Generation and Adaptive Reconstruction via MeanFlow for End-to-End Autonomous Driving
MeanFlow Identity 用平均速度场替代瞬时速度场，消除 ODE 求解误差并实现单步推理，这一思路可迁移到视觉生成任务（如扩散/流匹配加速）；GMN 提供连续轨迹表示，避免离散锚框词表，对需要多模态输出的规划问题有参考价值。
原文依据 MeanFlow Identity, which models the mean velocity field between GMN and data distribution instead of the instantaneous velocity field used in naive flow-matching methods
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论生成模型在轨迹规划中展现出巨大潜力。近期研究表明，锚框引导的生成模型能有效建模驾驶行为的不确定性并提升整体性能。然而，这些方法依赖离散锚框词表，测试时需充分覆盖轨迹分布以保证鲁棒性，导致词表大小与模型性能之间存在内在权衡。为克服这一局限，我们提出MeanFuser，一种通过三项关键设计提升效率与鲁棒性的端到端自动驾驶方法。（1）引入高斯混合噪声（GMN）引导生成采样，实现轨迹空间的连续表示，消除对离散锚框词表的依赖。（2）引入“MeanFlow Identity”，建模GMN与数据分布之间的平均速度场，而非朴素流匹配方法中的瞬时速度场，有效消除ODE求解器的数值误差并显著加速推理。（3）设计轻量级自适应重建模块（ARM），使模型能考虑所有采样提案，并在无提案令人满意时自适应决定是否重建轨迹。在NAVSIM闭环基准上的实验表明，MeanFuser取得了出色性能与卓越推理效率，为端到端自动驾驶提供了鲁棒且高效的解决方案。
英文原文摘要 1561 字符
Generative models have shown great potential in trajectory planning. Recent studies demonstrate that anchor-guided generative models are effective in modeling the uncertainty of driving behaviors and improving overall performance. However, these methods rely on discrete anchor vocabularies that must sufficiently cover the trajectory distribution during testing to ensure robustness, inducing an inherent trade-off between vocabulary size and model performance.To overcome this limitation, we propose MeanFuser, an end-to-end autonomous driving method that enhances both efficiency and robustness through three key designs. (1) We introduce Gaussian Mixture Noise (GMN) to guide generative sampling, enabling a continuous representation of the trajectory space and eliminating the dependency on discrete anchor vocabularies. (2) We introduce "MeanFlow Identity", which models the mean velocity field between GMN and data distribution instead of the instantaneous velocity field used in naive flow-matching methods, effectively eliminating numerical errors from ODE solvers and significantly accelerating inference. (3) We design a lightweight Adaptive Reconstruction Module (ARM) that enables the model to consider all sampled proposals and adaptively decide whether to reconstruct a trajectory when none of the proposals is satisfactory. Experiments on the NAVSIM closed-loop benchmark demonstrate that MeanFuser achieves outstanding performance and exceptional inference efficiency, offering a robust and efficient solution for end-to-end autonomous driving.