MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Xun, Zhao, Shijia, Wang, Yunxiang, Lu, Xin, Zhang, Wanfa, Qu, Rongsheng, Li, Weixin, Wang, Yunhong, Wen, Chenglu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911521468055552
author Huang, Xun
Zhao, Shijia
Wang, Yunxiang
Lu, Xin
Zhang, Wanfa
Qu, Rongsheng
Li, Weixin
Wang, Yunhong
Wen, Chenglu
author_facet Huang, Xun
Zhao, Shijia
Wang, Yunxiang
Lu, Xin
Zhang, Wanfa
Qu, Rongsheng
Li, Weixin
Wang, Yunhong
Wen, Chenglu
contents Embodied navigation is a fundamental capability for robotic agents operating. Real-world deployment requires open vocabulary generalization and low training overhead, motivating zero-shot methods rather than task-specific RL training. However, existing zero-shot methods that build explicit 3D scene graphs often compress rich visual observations into text-only relations, leading to high construction cost, irreversible loss of visual evidence, and constrained vocabularies. To address these limitations, we introduce the Multi-modal 3D Scene Graph (M3DSG), which preserves visual cues by replacing textual relational edges with dynamically assigned images. Built on M3DSG, we propose MSGNav, a zero-shot navigation system that includes a Key Subgraph Selection module for efficient reasoning, an Adaptive Vocabulary Update module for open vocabulary support, and a Closed-Loop Reasoning module for accurate exploration reasoning. Additionally, we further identify the last mile problem in zero-shot navigation determining the feasible target location with a suitable final viewpoint, and propose a Visibility-based Viewpoint Decision module to explicitly resolve it. Comprehensive experimental results demonstrate that MSGNav achieves state-of-the-art performance on the challenging GOAT-Bench and HM3D-ObjNav benchmark. The code will be publicly available at https://github.com/ylwhxht/MSGNav.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10376
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation
Huang, Xun
Zhao, Shijia
Wang, Yunxiang
Lu, Xin
Zhang, Wanfa
Qu, Rongsheng
Li, Weixin
Wang, Yunhong
Wen, Chenglu
Computer Vision and Pattern Recognition
Robotics
Embodied navigation is a fundamental capability for robotic agents operating. Real-world deployment requires open vocabulary generalization and low training overhead, motivating zero-shot methods rather than task-specific RL training. However, existing zero-shot methods that build explicit 3D scene graphs often compress rich visual observations into text-only relations, leading to high construction cost, irreversible loss of visual evidence, and constrained vocabularies. To address these limitations, we introduce the Multi-modal 3D Scene Graph (M3DSG), which preserves visual cues by replacing textual relational edges with dynamically assigned images. Built on M3DSG, we propose MSGNav, a zero-shot navigation system that includes a Key Subgraph Selection module for efficient reasoning, an Adaptive Vocabulary Update module for open vocabulary support, and a Closed-Loop Reasoning module for accurate exploration reasoning. Additionally, we further identify the last mile problem in zero-shot navigation determining the feasible target location with a suitable final viewpoint, and propose a Visibility-based Viewpoint Decision module to explicitly resolve it. Comprehensive experimental results demonstrate that MSGNav achieves state-of-the-art performance on the challenging GOAT-Bench and HM3D-ObjNav benchmark. The code will be publicly available at https://github.com/ylwhxht/MSGNav.
title MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2511.10376