MSGNav：释放多模态 3D 场景图在零样本具身导航中的潜力
MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation
用动态分配图像替代文本关系边构建多模态 3D 场景图，保留视觉证据以支持开放词表零样本导航；其关键子图选择与基于可见性的最终视点决策模块，可直接迁移到需要从 3D 场景表示中检索目标并确定可行停靠位姿的视觉任务。
原文依据 we introduce the Multi-modal 3D Scene Graph (M3DSG), which preserves visual cues by replacing textual relational edges with dynamically assigned images
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论具身导航是机器人智能体运行的一项基础能力。真实世界部署需要开放词表泛化能力与较低的训练开销，这促使研究者采用零样本方法，而非针对特定任务进行强化学习训练。然而，现有构建显式 3D 场景图的零样本方法往往将丰富的视觉观测压缩为纯文本关系，导致构建成本高、视觉证据不可逆丢失以及词表受限。为解决这些局限，我们提出多模态 3D 场景图（M3DSG），通过用动态分配的图像替代文本关系边来保留视觉线索。基于 M3DSG，我们提出 MSGNav，一个零样本导航系统，包含用于高效推理的关键子图选择模块、用于支持开放词表的自适应词表更新模块，以及用于准确探索推理的闭环推理模块。此外，我们进一步识别出零样本导航中的“最后一英里”问题——即确定具有合适最终视点的可行目标位置，并提出基于可见性的视点决策模块来显式解决该问题。大量实验结果表明，MSGNav 在 GOAT-Bench 与 HM3D-OVON 数据集上取得了最优性能。开源代码将公开发布。
英文原文摘要 1356 字符
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 GOAT-Bench and HM3D-OVON datasets. The open-source code will be publicly available.