# StageVLN：用于高效视觉语言导航的空间与轨迹辅助引导

> 原文链接：https://robotics.hellobike.com/papers/2026-10-05-2610-05664
> 频道：papers | 更新日期：2026-10-05 | 标签：arxiv, cs.RO
> 核心摘要：StageVLN通过训练期空间与轨迹辅助引导塑造VLN表征，并在R2R-CE和RxR-CE上验证无需推理期几何编码器。

# StageVLN：用于高效视觉语言导航的空间与轨迹辅助引导

::: v-pre
> 原文链接：[StageVLN: Spatial and Trajectory Auxiliary Guidance for Efficient Vision-Language Navigation](https://arxiv.org/abs/2610.05664)（arXiv 2610.05664 · 发布日期：2026-10-05）

**StageVLN: Spatial and Trajectory Auxiliary Guidance for Efficient Vision-Language Navigation**

StageVLN通过训练期空间与轨迹辅助引导塑造VLN表征，并在R2R-CE和RxR-CE上验证无需推理期几何编码器。

## 核心要点

- 标准动作监督难以显式保持场景几何、相对朝向和全局进度。
- 冻结几何基础模型提供多层空间引导，并加入相对航向与专家路线进度目标。
- 辅助组件仅训练时使用，部署时移除，保留原有推理路径。
- 在R2R-CE验证未见集达56.3% SR、51.4% SPL，在RxR-CE达54.3% SR。

**验证范围与局限**：验证范围为R2R-CE validation-unseen与RxR-CE，摘要未提及真机实验。

## 英文摘要 (Abstract)

Vision-and-Language Navigation (VLN) policies increasingly benefit from strong semantic priors provided by large vision-language models (VLMs). However, standard action supervision does not explicitly encourage intermediate representations to preserve scene geometry, relative orientation, or global episode progress. Incorporating depth estimators, explicit maps, point clouds, or geometry foundation models at inference can provide such structure but introduces additional computation, memory overhead, and architectural dependence during deployment. We introduce StageVLN, a training framework that shapes navigation representations through privileged spatial and trajectory guidance while preserving the original inference pathway. A frozen geometry foundation model provides multi-level spatial guidance to hierarchical navigator states, while relative-heading and expert-route progress objectives provide complementary trajectory-state supervision. All auxiliary components are used only during training and removed at deployment. On R2R-CE validation-unseen, StageVLN achieves 56.3\% SR and 51.4\% SPL with a 4B-parameter backbone, without an additional geometry encoder at inference. On RxR-CE, it achieves 54.3\% SR without additional navigation training data or a geometry encoder at inference.

作者：Anh Dao, Quan-Dung Pham, Le Danh Vinh, The Anh Nguyen, Nguyen Viet Tri Pham, Yiyu Chen, Pham Tuyen Le, Van‐Truong Nguyen, Quan Nguyen
:::

## 原始参考来源 (Sources)

- [arXiv 2610.05664：StageVLN: Spatial and Trajectory Auxiliary Guidance for Efficient Vision-Language Navigation](https://arxiv.org/abs/2610.05664)