FloVerse：户型图引导的多模态导航
FloVerse: Floor Plan-Guided Multi-Modal Navigation
把户型图作为紧凑空间先验注入具身导航策略，并用掩码模态建模训练扩散式多模态目标推理模块，可复用到任何需要跨模态目标定位与安全执行的视觉导航或VLA系统中。
原文依据 a diffusion-based multimodal goal-reasoning module trained via masked-modality modeling
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论户型图蕴含紧凑的空间先验，使智能体能够更高效地导航未见场景。尽管已有工作探索了户型图引导的导航，但主要集中于 PointNav 和有限的环境集合。为弥补这一空白，我们提出 FloVerse，一个面向户型图引导具身导航的新任务，统一了 PointNav、ObjectNav 与 ImageNav。为支撑 FloVerse，我们构建了 FloVerse-1.6K，一个包含来自 HM3D 与 Gibson 4+ 的 1.6K 个场景并配有对应户型图的大规模数据集，涵盖 24 万条专家轨迹与 1200 万帧 RGBD 图像。我们进一步提出 ThreeDiff，一种两阶段模仿学习策略，包含一个规划器——通过掩码模态建模训练的基于扩散的多模态目标推理模块，以及一个精化器——用于安全执行的基于深度的轨迹精化模块。大量实验表明：(1) 户型图先验在全部目标模态上均提升了导航性能；(2) ThreeDiff 隐式地从户型图中捕获空间信息。这些结果凸显了空间先验的有效性，并验证了我们所提出的统一方法在户型图引导具身导航中的价值。
英文原文摘要 1191 字符
Floor plans encapsulate compact spatial priors, enabling agents to navigate unseen scenes more efficiently. While prior work has explored floor plan-guided navigation, it has focused mainly on PointNav and a limited set of environments. To bridge this gap, we introduce FloVerse, a new task for floor plan-guided embodied navigation that unifies PointNav, ObjectNav, and ImageNav. To support this FloVerse, we assemble FloVerse-1.6K, a large-scale dataset of 1.6K scenes from HM3D and Gibson 4+, paired with corresponding floor plans, comprising 240K expert trajectories and 12M RGBD frames. We further propose ThreeDiff, a two-stage imitation learning policy comprising a planner, a diffusion-based multimodal goal-reasoning module trained via masked-modality modeling, and a refiner, a depth-based trajectory-refinement module for safe execution. Extensive experiments demonstrate that (1) floor-plan priors improve navigation performance across all goal modalities, and (2) ThreeDiff implicitly captures spatial information from floor plans. These results underscore the effectiveness of spatial priors and validate our proposed unified approach for floor plan-guided embodied navigation.