English(EN) SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gating

SafeFlow框架支持实时、物理引导的人形机器人控制

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-16 04:00

研究人员开发了SafeFlow，一个用于实时文本驱动的人形机器人控制的新框架。该系统集成了物理引导的运动生成和多级安全门控，以确保生成的轨迹在物理上可行且对现实世界执行安全。SafeFlow利用变分自编码器潜在空间和校正流匹配方法，通过Reflow加速，来生成运动计划。其安全机制包括检测分布外提示、过滤不稳定生成以及在Unitree G1等机器人上执行前强制执行硬运动学约束。 AI

影响增强了人形机器人的安全性和实时控制能力，可能实现更复杂、更可靠的人机交互。

排序理由该集群包含一篇详细介绍机器人新AI驱动控制框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 16 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍机器人新AI驱动控制框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, product, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Hanbyel Cho, Sang-Hun Kim, Jeonguk Kang, Donghan Koo · 2026-09-16 04:00

SafeFlow：通过物理引导的校正流和选择性安全门控实现文本驱动的实时人形全身控制

arXiv:2603.23983v2 Announce Type: replace-cross Abstract: Recent advances in real-time interactive text-driven motion generation have enabled humanoids to perform diverse behaviors. However, kinematics-only generators often exhibit physical hallucinations, producing motion trajec…

本文如何被排名

报道来源 [1]

SafeFlow：通过物理引导的校正流和选择性安全门控实现文本驱动的实时人形全身控制

相关实体

相关话题