---
格式版本: 2
标题: "Edge-Native Embodied Intelligence for Action-Aware Wireless Edge Networks"
原文链接: "https://arxiv.org/abs/2608.17774"
发布日期: "2026-08-18"
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\\]** Tue, 18 Aug 2026 13:39:12 UTC (2,955 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: "2026-08-20T15:38:15+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: "2026-08-20T15:33:55+08:00"
入库时间: "2026-08-20T07:38:15.657Z"
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=latency&searchtype=all"
匹配关键词:
  - "latency"
  - "deployment"
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优质: "否"
AI打分: 6
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "文章为arXiv搜索入口页，命中关键词latency但实际内容为无线边缘网络具身智能论文，与超节点、AI Rack、机柜级AI基础设施、供电散热互连等主题完全无关，判定为无关资料。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T15:41:08+08:00"
AI主题相关性: 0
AI来源权威性: 5
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 1
AI摘要: "该文提出边缘原生具身智能（ENEI）框架，将具身代理、6G通信网络和边缘认知服务整合为双向边缘—具身循环，解决具身智能在资源受限环境下的物理接地与行动感知不足。案例验证其在OOD无人机导航和移动联邦学习中可行且通信高效。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:21:42.732Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.17774"
---

## Electrical Engineering and Systems Science > Systems and Control

## Title:Edge-Native Embodied Intelligence for Action-Aware Wireless Edge Networks

Authors:[Yiru Wang](https://arxiv.org/search/eess?searchtype=author&query=Wang,+Y), [Chuanao Jiang](https://arxiv.org/search/eess?searchtype=author&query=Jiang,+C), [Jiahui Cui](https://arxiv.org/search/eess?searchtype=author&query=Cui,+J), [Zide Fan](https://arxiv.org/search/eess?searchtype=author&query=Fan,+Z), [Lei Wang](https://arxiv.org/search/eess?searchtype=author&query=Wang,+L), [Zehui Xiong](https://arxiv.org/search/eess?searchtype=author&query=Xiong,+Z), [Dong In Kim](https://arxiv.org/search/eess?searchtype=author&query=Kim,+D+I)

[View PDF](https://arxiv.org/pdf/2608.17774) [HTML (experimental)](https://arxiv.org/html/2608.17774v1)

> Abstract:Embodied intelligence is shifting artificial intelligence from passive digital perception toward active physical interaction. However, foundation-model-enabled embodied agents face a fundamental tension between open-world cognition and resource-constrained deployment. On-device models are limited by computation, memory, and energy budgets, whereas cloud-centric solutions introduce latency and reliability risks over dynamic wireless links. Edge general intelligence provides a promising cognitive backbone, but existing frameworks still lack physical grounding, action awareness, and mechanisms for actively acquiring useful physical experience. To address these limitations, this article introduces edge-native embodied intelligence (ENEI), an action-aware wireless edge framework that integrates embodied agents, the 6G communication and networking fabric, and edge cognitive services into a 6G-mediated bidirectional edge-embodiment loop. Along the edge-to-embodiment axis, confidence-aware assistance and edge-driven generative adaptation enhance local autonomy under out-of-distribution (OOD) conditions. Along the embodiment-to-edge axis, value-of-experience guided active embodied federated learning enables physical actions to generate informative experience for continuous edge model evolution. The 6G fabric supports both directions through goal-oriented transmission and programmable radio-resource allocation. Two case studies on OOD drone navigation and mobility-driven federated learning illustrate the feasibility and communication efficiency of the proposed mechanisms. ENEI provides a unified perspective in which edge cognition strengthens embodied action, while embodied agency actively enriches edge cognition, laying the foundation for scalable, adaptive, and self-evolving embodied wireless systems.

| Subjects: | Systems and Control (eess.SY); Signal Processing (eess.SP) |
| --- | --- |
| Cite as: | [arXiv:2608.17774](https://arxiv.org/abs/2608.17774) \[eess.SY\] |
|  | (or [arXiv:2608.17774v1](https://arxiv.org/abs/2608.17774v1) \[eess.SY\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.17774](https://doi.org/10.48550/arXiv.2608.17774) |

## Submission history

From: Yiru Wang \[[view email](https://arxiv.org/show-email/799b6eb0/2608.17774)\]  
**\[v1\]** Tue, 18 Aug 2026 13:39:12 UTC (2,955 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.17774) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))