eaiadmin
ccf72e32bd
包含6个文件： - 总览：7条关键发现摘要 - 国内同行：翼辉SylixOS、华为鸿蒙/LiteOS、端侧AI框架 - 国际同行：FreeRTOS/Zephyr/ThreadX/VxWorks对比、LLM运行时框架、Agent框架、NVIDIA Jetson物理AI - 学术文献：14篇论文/文章完整摘要和链接 - 关键研究空白：7个研究空白及与项目框架的对齐 - 文献索引：40+条文献完整列表
75 lines
6.9 KiB
Markdown
75 lines
6.9 KiB
Markdown
# 05-文献索引
|
|
|
|
## 完整文献列表
|
|
|
|
### 学术论文（arXiv + 期刊）
|
|
|
|
| # | 论文标题 | 作者 | 年份 | 来源 | 链接 |
|
|
|---|---------|------|------|------|------|
|
|
| 1 | Toward a Modular Architecture for Embedded Agent Systems at the Edge | Marcus Rüb, Michael Gerhards | 2026 | arXiv:2606.02862 | [链接](https://arxiv.org/html/2606.02862) |
|
|
| 2 | Real-time Operating Systems (RTOS) For Edge AI | Hasan, Chakawuya, Razaul | 2025 | Int'l Journal of Trends in Emerging Research and Development | [PDF](https://www.researchtrendsjournal.com/uploads/articles/3-4-38.1.pdf) |
|
|
| 3 | Agent.xpu: Efficient Scheduling of Agentic LLM Workloads on Heterogeneous SoC | Wei, Zhang, Li等 | 2026 | arXiv:2506.24045 | [链接](https://arxiv.org/html/2506.24045v2) |
|
|
| 4 | LLM Inference at the Edge: Mobile, NPU, and GPU Performance Trade-offs | Tummalapalli, Arayakandy, Pal, Kundan | 2026 | arXiv:2603.23640 | [链接](https://arxiv.org/html/2603.23640v2) |
|
|
| 5 | Efficient Inference for Edge Large Language Models: A Survey | Cai, Tian, Yang, Jia, Li, Wang | 2025 | Science China Information Sciences | [PDF](https://www.sciopen.com/local/article_pdf/10.26599/TST.2025.9010166.pdf) |
|
|
| 6 | Edge-Centric Generative AI: A Survey on Efficient Inference for LLMs in Resource-Constrained Environments | Rui Huang | 2026 | Journal of Computer and Communications 14(4) | [PDF](https://content.scirp.org/pdf/jcc_1733489.pdf) |
|
|
| 7 | LLM Inference Scheduling: A Survey of Techniques, Frameworks, and Trade-offs | 华为技术团队 (Zhang, Fan等) | 2025 | TechRxiv | [PDF](https://www.techrxiv.org/doi/pdf/10.36227/techrxiv.176238087.79673350/v1?download=true) |
|
|
|
|
### 行业文章/博客
|
|
|
|
| # | 标题 | 作者/机构 | 年份 | 来源 | 链接 |
|
|
|---|------|----------|------|------|------|
|
|
| 8 | 翼辉SylixOS:中国自主硬实时操作系统的崛起之路 | CSDN博主 | 2026 | CSDN | [链接](https://blog.csdn.net/hujunming/article/details/148370364) |
|
|
| 9 | 翼辉信息SylixOS AI应用方案发布 | 翼辉信息 | 2025 | 电子发烧友 | [链接](https://www.elecfans.com/d/6800705.html) |
|
|
| 10 | 基于AI辅助工程的混合MCU-边缘-云平台的工业AIoT系统设计 | 吴薇 | 2026 | Hans Publishers | [PDF](https://pdf.hanspub.org/etis_1380065.pdf) |
|
|
| 11 | 鸿蒙开发之路：端侧模型压缩、量化与加速技术详解 | zq18 | 2026 | 博客园 | [链接](https://www.cnblogs.com/zq18/p/19263508) |
|
|
| 12 | 鸿蒙AI框架(HiAI/MindSpore Lite)集成与推理优化 | bug菌 | 2025 | 华为云社区 | [链接](https://bbs.huaweicloud.cn/blogs/467036) |
|
|
| 13 | 端侧 AI 框架让智能真正贴身随行 | 佩臻 | 2026 | 华为开发者联盟 | [链接](https://developer.huawei.com/consumer/cn/blog/topic/03203187067519385) |
|
|
| 14 | HiSilicon Kirin A1互联穿戴生态设备 | CSDN博主 | 2025 | CSDN | [链接](https://blog.csdn.net/weixin_33193177/article/details/154404675) |
|
|
| 15 | MCU跑AI的RTOS选型:FreeRTOS、ThreadX与Zephyr对比 | 匿名 | 2026 | CSDN | [链接](https://blog.csdn.net/weixin_30036289/article/details/164544991) |
|
|
| 16 | Unlock Faster, Smarter Edge Models with 7x Gen AI Performance on Jetson AGX Thor | NVIDIA技术博客 | 2025 | NVIDIA Developer | [链接](https://developer.nvidia.com/blog/unlock-faster-smarter-edge-models-with-7x-gen-ai-performance-on-nvidia-jetson-agx-thor) |
|
|
| 17 | The Edge LLM Runtime Stack 2026 | Edge AI Stack | 2026 | edgeaistack.ai | [链接](https://edgeaistack.ai/blog/edge-llm-runtime-stack-2026/) |
|
|
| 18 | vLLM vs TensorRT-LLM: Inference Runtime Guide | Jérôme Lafont | 2026 | Cosmo Edge | [链接](https://cosmo-edge.com/vllm-vs-tensorrt-llm/) |
|
|
| 19 | Running LLMs on Jetson Orin, llama.cpp, Ollama | Aaron Angulo | 2026 | Medium | [链接](https://www.proventusnova.com/blog/llm-inference-jetson-orin-llamacpp-ollama/) |
|
|
| 20 | NVIDIA Jetson 将智能体 AI 带入物理世界 | NVIDIA中国 | 2026 | NVIDIA博客 | [链接](https://blogs.nvidia.cn/blog/jetson-agentic-ai-physical-world/) |
|
|
| 21 | NVIDIA Cosmos 3 Edge and the Robot AI Revolution 2026 | Arjun Mehta | 2026 | JustLast.in | [链接](https://www.justlast.in/nvidia-cosmos-3-edge-and-the-robot-ai-revolution-2026-on-device-world-models-humanoid-surgery-and-japans-2-3-billion-noetra-project/) |
|
|
| 22 | Edge AI for real-time robot control: NVIDIA Jetson Orin | SCALE D2C | 2026 | scaled2c.com | [链接](https://scaled2c.com/blog/physical-ai-robotics/edge-ai-for-real-time-robot-control-nvidia-jetson-orin.html) |
|
|
| 23 | Best AI Agent Frameworks (2026年7月) | AI Wiki | 2026 | aiwiki.ai | [链接](https://aiwiki.ai/wiki/best_ai_agent_frameworks) |
|
|
| 24 | LangChain vs CrewAI vs AutoGen: Which Agent Framework Scales? | Mark | 2026 | markaicode.com | [链接](https://markaicode.com/best/best-ai-agent-framework/) |
|
|
| 25 | Best Multi-Agent AI Frameworks for 2025 & 2026 | Datawhale | 2025 | langcopilot.com | [链接](https://langcopilot.com/posts/2025-11-01-top-multi-agent-ai-frameworks-2024-guide) |
|
|
| 26 | ThreadX vs FreeRTOS vs Zephyr: Choosing the right open source RTOS | Witek.io | 2026 | witekio.com | [链接](https://witekio.com/blog/threadx-freertos-zephyr-rtos/) |
|
|
| 27 | How to Choose the Best RTOS for Embedded Systems | VxWorks7 | 2026 | vxworks7.com | [链接](https://www.vxworks7.com/training/how-to-choose-the-best-rtos-for-embedded-systems/) |
|
|
| 28 | LiteOS-A内核技术全解析:架构、性能与开发实践 | 鸿蒙小白龙 | 2025 | jishuzhan.net | [链接](https://jishuzhan.net/article/1981610278161809409) |
|
|
|
|
### 产品文档/官方资料
|
|
|
|
| # | 标题 | 来源 | 链接 |
|
|
|---|------|------|------|
|
|
| 29 | SylixOS 大型实时操作系统文档中心 | 翼辉信息 | [链接](https://docs.acoinfo.com/sylixos/) |
|
|
| 30 | SylixOS 大型实时操作系统概述 | 人人都懂物联网 | [链接](https://getiot.tech/dictionary/sylixos/) |
|
|
| 31 | NVIDIA Jetson 平台概述 | NVIDIA | [链接](https://www.nvidia.com/en-us/autonomous-machines/) |
|
|
| 32 | Jetson Workshop GTC DC 2025 | Jetson AI Lab | [链接](https://www.jetson-ai-lab.com/workshop_gtcdc2025.html) |
|
|
| 33 | Running LLMs on Jetson Orin | jetson-containers (dusty-nv) | [链接](https://github.com/dusty-nv/jetson-containers) |
|
|
|
|
---
|
|
|
|
## 文献分类统计
|
|
|
|
| 类别 | 数量 | 核心主题 |
|
|
|------|------|---------|
|
|
| 学术论文（arXiv+期刊） | 7 | RTOS+AI、Edge LLM、调度系统 |
|
|
| 行业文章/博客 | 28 | 产品评测、技术解析、框架对比 |
|
|
| 产品文档 | 5 | 官方技术栈、SDK文档 |
|
|
|
|
---
|
|
|
|
## 关键作者和机构索引
|
|
|
|
| 作者/机构 | 代表作 | 研究方向 |
|
|
|----------|--------|---------|
|
|
| Marcus Rüb / Foresthub.Ai | arXiv:2606.02862 | 嵌入式Agent架构 |
|
|
| Xinming Wei / 北京大学 | arXiv:2506.24045 | Agent.xpu调度系统 |
|
|
| Hasan et al. / 西部师大 | RTOS for Edge AI (2025) | RTOS内核实证对比 |
|
|
| NVIDIA技术团队 | Jetson AI多篇 | 物理AI、边缘LLM |
|
|
| 华为技术团队 | LLM Inference Scheduling Survey | 推理调度综述 |
|
|
| 吴薇 / 杭州电子科技大学 | AI辅助工程方法 | MCU-边缘-云架构 |
|