研究人员开发了一个“语义自主栈”，使室内移动机器人能够理解自然语言指令，克服了当前视觉语言模型（VLM）的延迟和内存限制。该框架采用混合方法，由确定性解析器快速处理大多数指令，仅将模糊情况升级给VLM。一种新颖的语义记忆系统允许跨会话学习和机器人之间的知识转移，显著减少处理时间，并支持在Raspberry Pi 5等低功耗硬件上运行，无需板载GPU。 AI
影响 该框架有望在资源受限的设备上实现更直观的人机交互室内导航任务。
排序理由 这是一篇详细介绍机器人导航和自然语言理解新框架的研究论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2605.02525v1 Announce Type: cross Abstract: Autonomous indoor mobile robots can navigate reliably to metric coordinates using established frameworks such as ROS 2 Navigation 2, yet they lack the ability to interpret natural language instructions that express intent rather t…
Autonomous indoor mobile robots can navigate reliably to metric coordinates using established frameworks such as ROS 2 Navigation 2, yet they lack the ability to interpret natural language instructions that express intent rather than positions. Vision-Language Models offer the se…
Autonomous indoor mobile robots can navigate reliably to metric coordinates using established frameworks such as ROS 2 Navigation 2, yet they lack the ability to interpret natural language instructions that express intent rather than positions. Vision-Language Models offer the se…