论文/arXiv 2610.10635·21 天前

说导航器的语言：面向冻结空中 VLN 智能体的轨迹扎根指令翻译

针对空中VLN用户短指令与训练命令不匹配，TGIT从轨迹结果学习翻译，并在OpenFly、人类指令及CityNav/AirVLN上验证。

AI 整理

• 本体 SDK 与控制
• 感知与导航

原文链接：Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents（arXiv 2610.10635 · 发布日期：2026-10-07）

Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents

针对空中VLN用户短指令与训练命令不匹配，TGIT从轨迹结果学习翻译，并在OpenFly、人类指令及CityNav/AirVLN上验证。

核心要点

• 空中VLN常用细节丰富且轨迹对齐的命令训练，但用户多给出短而意图驱动的指令。
• 作者用人写风格示例提示语言模型，将原始命令转成成对的意图中心Weak命令以扩展训练。
• TGIT作为前端保持导航器冻结，依据其轨迹结果学习把Weak输入翻译为可执行命令。
• Weak训练翻译器提升Weak输入成功率，并零样本迁移到真实人类指令及其他环境。

验证范围与局限：摘要未说明仿真或真机及机器数量；验证范围包括OpenFly、真实人类指令、留出OpenFly、CityNav和AirVLN。

英文摘要 (Abstract)

Aerial vision-and-language navigation (VLN) agents are typically trained on detail-rich, trajectory-aligned commands, whereas users issue short, intent-driven instructions; on a frozen OpenFly navigator, this \emph{instruction gap} drops success rate (SR) from $31.03%$ to $11.33%$. To scale translator training, we prompt a language model with human-written style examples to convert original commands into paired, intent-centered Weak commands, which yield $15.27%$ SR. We introduce the \textbf{Trajectory-Grounded Instruction Translator (TGIT)}, a front-end that keeps the navigator frozen and translates Weak inputs into agent-executable commands by learning from its trajectory outcomes. The resulting Weak-trained translator raises Weak-input SR to $37.93%$ and transfers zero-shot to real human instructions ($11.33%{\rightarrow}32.51%$); it also improves held-out OpenFly ($4.95%{\rightarrow}20.79%$) and yields recovery on CityNav and AirVLN.

作者：Xi Chen, Zhe Liu, Xiaogang Xu, Jiafei Xu, Chunyi Zhou, Yuan Su, Rui Zeng, Tianyu Du, Kelu Yao, Chao Li

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.10635：Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents访问 2026-10-21