arXiv cs.AI· Zenghui Yang, Xinan Wang, Danyang Li, Zhe Xu, Zheng Yang, Jingao Xu, Qiang Ma·· 3 小时前AI 评分34
EMS-Nav：面向具身导航的多尺度框架
EMS-Nav: Embodied Multi-Scale Navigation
AI 导读
EMS-Nav 是一个受生物启发的多尺度具身导航框架，以区域为基本单元组织感知、空间理解与导航，支持端侧在线运行。在物体、房间、区域和特定实例四类目标的多尺度导航实验中，EMS-Nav 首次尝试成功率达最高，比最强基线整体提升 9.39 个百分点，同时降低建图开销与语言模型输入规模。真实室内环境的本地部署进一步验证了其在资源受限机器人上的可用性。
正文 · 原文
Abstract:Industrial inspection robots must reach work areas and specific equipment. These destinations span spatial scales from regions to objects, so robots must relate each target to its surroundings to interpret the task and determine where to go. Existing methods support multi-scale navigation through rich environment representations, but at substantial computational and storage cost. These costs limit deployment where wireless coverage is incomplete and robots must rely on onboard computation to keep navigating. We introduce EMS-Nav, a biologically inspired multi-scale navigation framework for online, on-device operation. Its core innovation is to organize perception, spatial understanding, and navigation around regions, making local spaces the basic units of both environmental understanding and computation. EMS-Nav groups objects and their relationships into functionally meaningful spaces and builds a hierarchy from local areas to the broader environment. This organization supports efficient map construction alongside the interpretation and execution of language goals at different scales. In multi-scale navigation experiments, EMS-Nav achieves the highest success rates across all four target categories: objects, rooms, regions, and specific instances. It outperforms state-of-the-art open-vocabulary mapping and navigation methods, improving overall first-attempt success by 9.39 percentage points over the strongest baseline while reducing mapping overhead and language-model input size. Local deployment in real indoor environments further demonstrates its applicability to on-device embodied navigation. By establishing regions as the organizing basis for perception, spatial understanding, and language-guided action, EMS-Nav provides a system that combines multi-scale understanding with computational efficiency to support diverse language goals on resource-constrained robots.
| Comments: | 30 pages, including appendices |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2611.15589 [cs.RO] |
| (or arXiv:2611.15589v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15589 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zenghui Yang [view email]
[v1]
Wed, 11 Nov 2026 15:09:28 UTC (3,253 KB)
来源：arXiv cs.AI · arxiv.org