Scholarlink

Abstract <p>Cross-floor robotic navigation has become an increasingly critical capability for autonomous systems operating in multi-floor buildings. While 3D scene graphs have demonstrated promise for representing hierarchical spatial relationships, current approaches predominantly address cross-floor navigation by stairs, overlooking the practical challenges of elevator-mediated navigation in modern buildings. This paper presents <b>ElevNav</b>, a novel framework that bridges this gap through two key innovations: (1) automatic construction of semantically-rich 3D scene graphs from RGB-D sequences with estimated camera trajectories, and (2) task decomposition using large language models to translate natural language commands into executable action sequences. Our method addresses elevator interaction through specialized action primitives such as pressing buttons, entering and exiting the elevator, and moving toward target objects. We evaluate ElevNav in complex simulated environments built using Isaac Sim, demonstrating robust performance in multi-floor navigation scenarios. To facilitate further research, we release a new dataset containing elevator environments with corresponding scene graph representations, addressing a critical gap in existing 3D navigation benchmarks, which is open-sourced at: <a href="https://github.com/zhanghuzhenyu/elevnav">https://github.com/zhanghuzhenyu/elevnav</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

ElevNav: Large Language Model-Guided Robot Navigation via 3D Scene Graphs in Elevator Environments

  • 期刊论文

  • 发表日期：

    2025年9月17日

查看全文
  • Huzhenyu Zhang

摘要

Abstract

Cross-floor robotic navigation has become an increasingly critical capability for autonomous systems operating in multi-floor buildings. While 3D scene graphs have demonstrated promise for representing hierarchical spatial relationships, current approaches predominantly address cross-floor navigation by stairs, overlooking the practical challenges of elevator-mediated navigation in modern buildings. This paper presents ElevNav, a novel framework that bridges this gap through two key innovations: (1) automatic construction of semantically-rich 3D scene graphs from RGB-D sequences with estimated camera trajectories, and (2) task decomposition using large language models to translate natural language commands into executable action sequences. Our method addresses elevator interaction through specialized action primitives such as pressing buttons, entering and exiting the elevator, and moving toward target objects. We evaluate ElevNav in complex simulated environments built using Isaac Sim, demonstrating robust performance in multi-floor navigation scenarios. To facilitate further research, we release a new dataset containing elevator environments with corresponding scene graph representations, addressing a critical gap in existing 3D navigation benchmarks, which is open-sourced at: https://github.com/zhanghuzhenyu/elevnav.