论文/arXiv 2609.12285·1 个月前

AnchorVLN：面向开放词汇导航的几何锚定视觉-语言定位推理

AnchorVLN以VLM提出语义、几何决定度量，通过MCP工具在CMU VLN Challenge 2026两类任务上验证开放词汇导航。

AI 整理

• 本体 SDK 与控制
• 感知与导航
• 具身模型

原文链接：AnchorVLN: Geometry-Anchored Vision-Language Grounding Reasoning for Open-Vocabulary Navigation（arXiv 2609.12285 · 发布日期：2026-09-10）

AnchorVLN: Geometry-Anchored Vision-Language Grounding Reasoning for Open-Vocabulary Navigation

AnchorVLN以VLM提出语义、几何决定度量，通过MCP工具在CMU VLN Challenge 2026两类任务上验证开放词汇导航。

核心要点

• 问题：VLM具备开放词汇定位与零样本推理，但难以直接从图像输出可靠度量量。
• 方法：VLM只提出语义，几何负责距离、方位和比较空间关系等度量判断。
• 系统实现为EMBODIED-NAV-MCP，用可调用工具约束语义与几何边界。
• 实验：指令跟随达64.4%；物体指代10/45越过重叠阈值，优于直接坐标估计。

验证范围与局限：验证范围为CMU Vision-Language Navigation Challenge 2026的15个场景30个指令跟随问题和45个物体指代问题。

英文摘要 (Abstract)

Vision-Language Navigation (VLN) in unseen indoor environments is useful in real-world robotics, where an agent must follow natural-language instructions, locate objects, and answer spatial questions without a pre-built map or fixed object vocabulary. Multimodal vision-language models (VLMs) provide strong open-vocabulary grounding and zero-shot reasoning, but struggle to emit reliable metric quantities such as range, bearing, and comparative spatial relations directly from images. Existing approaches address this by folding geometry into hand-engineered pipelines or asking models to output waypoints, requiring changes to the control stack for different robots, tasks, or vocabularies. We introduce AnchorVLN, an open-vocabulary VLN system built on a simple rule: the VLM proposes semantics; geometry decides metrics. It is realised as EMBODIED-NAV-MCP, a Model Context Protocol (MCP) server driven by a VLM agent through a compact set of callable tools. Since no tool accepts distance in metres or bearing in radians, the schema enforces the semantic-geometry boundary without modifying the downstream autonomy stack. We benchmark both tasks of the CMU Vision-Language Navigation Challenge 2026: 30 instruction-following questions over 15 scenes and a frozen 45-question object-reference set. The full system achieves 64.4 percent on instruction following, dropping by 13.3 percentage points without controller modeling (t = 2.77). On object reference, geometric anchoring clears the challenge overlap threshold on 10 of 45 questions, versus 0 of 45 for direct coordinate estimation, reducing median center error from 3.37 m to 2.48 m.

作者：Long Giang Vu, Chenxu Yao, Yuxin Liu, FNU Aryan, Rajath C Aralikatti

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2609.12285：AnchorVLN: Geometry-Anchored Vision-Language Grounding Reasoning for Open-Vocabulary Navigation访问 2026-10-21