发表机构
Korea University of Technology and Education（韩国技术教育大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
DAWN通过将去噪和对比对齐融入世界模型，实现无需手动滤波器校准的噪声鲁棒四足跑酷，在Unitree Go1上零样本完成楼梯、间隙和台阶等复杂地形。
AI 中文摘要
基于视觉的腿足运动方法在训练时假设深度数据是干净的，并在部署时依赖手工调整的后处理滤波器。然而，滤波器参数很少公开，阻碍了可复现性，并且当深度噪声未被处理时，性能会大幅下降。将噪声鲁棒性直接构建到学习流程中可以消除这种依赖。虽然这种鲁棒性已在本体感觉输入方面得到探索，但在腿足运动中，针对深度感知的类似方法仍然缺乏。我们提出了DAWN（Denoising and Alignment in World models for Noise-robustness），一种用于腿足运动的噪声鲁棒感知框架，它通过两项修改将噪声鲁棒性直接构建到世界模型中：（1）将带噪深度输入编码器，同时将干净深度作为重建目标，迫使模型隐式地去噪其输入；（2）应用对比学习来对齐带噪和干净深度的潜在状态。重要的是，DAWN不依赖于特定的噪声模型，在部署时无需针对噪声分布进行手动调整。此外，与现有的基于世界模型的方法相比，它不产生额外的推理成本。在没有任何手动滤波器校准的情况下——仅依靠学习到的噪声鲁棒表示——DAWN在Unitree Go1上实现了零样本四足跑酷：从原始深度观测中穿越高达18厘米的楼梯、跨越高达70厘米的间隙以及登上高达45厘米的台阶。消融研究表明，去噪和对比对齐分别在重建和表示层面互补地贡献，并且结合时产生加性增益。视频和代码可在以下网址获取：this https URL
英文摘要
Vision-based legged locomotion methods assume clean depth at training time and rely on hand-tuned post-processing filters at deployment. However, filter parameters are rarely disclosed, hindering reproducibility, and performance degrades substantially when depth noise is left unaddressed. Building noise robustness directly into the learning pipeline would eliminate this dependency. While such robustness has been explored for proprioceptive inputs, analogous approaches for depth perception remain largely absent in legged locomotion. We propose DAWN (Denoising and Alignment in World models for Noise-robustness), a noise-robust perception framework for legged locomotion, which builds noise robustness directly into a world model via two modifications: (1) feeding noisy depth to the encoder while keeping clean depth as the reconstruction target, forcing the model to implicitly denoise its input; and (2) applying contrastive learning to align the latent states of noisy and clean depth. Importantly, DAWN is not tied to a specific noise model, requiring no manual tuning to the noise distribution at deployment. Furthermore, it incurs no additional inference cost over existing world model-based methods. Without any manual filter calibration -- relying solely on the learned noise-robust representation -- DAWN achieves zero-shot quadruped parkour on a Unitree Go1: traversing stairs up to 18 cm, clearing gaps up to 70 cm, and mounting steps up to 45 cm from raw depth observations. Ablation studies show that denoising and contrastive alignment contribute at complementary levels -- reconstruction and representation, respectively -- and yield additive gains when combined. Videos and code are available at: https://dawn-parkour.github.io/
Comments8 pages, 6 figures. Accepted to IROS 2026
