---
格式版本: 2
标题: "D3D-GEN: Robot-Aware Domain-Grounded Interactive 3D World Generation for Social Robotics"
原文链接: "https://arxiv.org/abs/2608.11876"
发布日期: "2026-08-12"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Wed, 12 Aug 2026 10:03:01 UTC (3,464 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-13T18:34:08+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-13T18:24:32+08:00"
入库时间: "2026-08-13T10:34:08.553Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=AI&searchtype=all"
匹配关键词:
  - "AI"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI质检状态: "评分失败"
AI评分尝试次数: 1
AI评分错误类型: "service_error"
AI评分错误: "LLM call failed; tried model chain: ali-deepseek-v4-flash -> tx-deepseek-v4-flash | Model ali-deepseek-v4-flash failed 500: {\"error\":{\"code\":\"\",\"message\":\"Database error, please contact the administrator (request id: 202608131034104050648798268d9d670WARyHS)\",\"type\":\"new_api_error\"}} | Model tx-deepseek-v4-flash failed 500: {\"error\":{\"code\":\"\",\"message\":\"Database error, please contact the administrator (request id: 202608131034117884400578268d9d6xDbSt1L1)\",\"type\":\"new_api_error\"}}"
AI评分开始时间: "2026-08-13T10:34:08.571Z"
AI评分结束时间: "2026-08-13T10:34:12.147Z"
AI摘要: "D3D-GEN是一个面向社交机器人仿真训练的三维世界生成系统，将领域智能体与检索增强生成（RAG）流程结合，从公开数据自动构建领域数据库，并据此生成可交互的floorplan和物体布局，不依赖固定3D模型库。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:40:18.613Z"
采集批次: "2026年8月13日18点24分27秒"
采集批次ID: "20260813-182427-1564e572"
去重键: "https://arxiv.org/abs/2608.11876"
---

## Computer Science > Robotics

## Title:D3D-GEN: Robot-Aware Domain-Grounded Interactive 3D World Generation for Social Robotics

[View PDF](https://arxiv.org/pdf/2608.11876) [HTML (experimental)](https://arxiv.org/html/2608.11876v1)

> Abstract:Training and validation of Embodied AI for social navigation critically depends on realistic simulation environments, yet many current approaches fail to find a balance between realism and simulability. We propose D3D-GEN, a novel world generation system that combines a domain agent with a retrieval-augmented generation (RAG) pipeline grounded in that domain. Our system enables users to rapidly generate domain-grounded, fully interactive 3D worlds by automating both the collection of domain knowledge and the synthesis of realistic floorplans and object placements, without dependence on any fixed 3D model database. Given a domain description prompt, the research agent collects publicly accessible domain-specific data and constructs a persistent domain database. Using this database, our RAG pipeline generates plausible floorplans and object placements by dynamically querying a user-provided semantic database, which can be easily extended or modified. The output is a fully interactive 3D world loadable by the popular simulators Isaac Sim and Gazebo. With our approach, we have built databases for several common domains (indoor residential, hospital, office) and generated dozens of distinct, plausible simulation environments for each domain. We present D3D-GEN with a local web frontend that facilitates rapid, interactive world generation for robot simulation.

| Comments: |  |
| --- | --- |
| Subjects: | Robotics (cs.RO); Human-Computer Interaction (cs.HC) |
| MSC classes: | 68T40 |
| ACM classes: | I.2.9; I.3.5 |
| Cite as: | [arXiv:2608.11876](https://arxiv.org/abs/2608.11876) \[cs.RO\] |
|  | (or [arXiv:2608.11876v1](https://arxiv.org/abs/2608.11876v1) \[cs.RO\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.11876](https://doi.org/10.48550/arXiv.2608.11876) |

## Submission history

From: Anh Do Duc \[[view email](https://arxiv.org/show-email/b054744b/2608.11876)\]  
**\[v1\]** Wed, 12 Aug 2026 10:03:01 UTC (3,464 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.11876) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))