让机器人更懂指令，实现全身协调的物理感知运动控制。
PhysiFlow: Physics-Aware Humanoid Whole-Body VLA via Multi-Brain Latent Flow Matching and Robust Tracking
- 通过多脑隐空间流匹配，融合视觉、语言与动作意图
- 在动态任务中实现全身协调，提升运动稳定性
- 适合需要语义引导的复杂人形机器人控制场景
在人形机器人控制领域，视觉-语言-动作（VLA）与全身控制的融合对语义引导的真实世界任务执行至关重要。然而，现有方法存在推理效率低或缺乏有效语义引导的问题，导致动态肢体协调任务中出现不稳定性。为此，我们提出一种语义-运动意图引导、具备物理感知能力的多脑VLA框架，用于人形机器人全身控制。一系列实验评估了所提框架的性能，结果表明该框架实现了可靠、视觉-语言引导的全身协同运动控制。
原文摘要 · Abstract (English)
In the domain of humanoid robot control, the fusion of Vision-Language-Action (VLA) with whole-body control is essential for semantically guided execution of real-world tasks. However, existing methods encounter challenges in terms of low VLA inference efficiency or an absence of effective semantic guidance for whole-body control, resulting in instability in dynamic limb-coordinated tasks. To bridge this gap, we present a semantic-motion intent guided, physics-aware multi-brain VLA framework for humanoid whole-body control. A series of experiments was conducted to evaluate the performance of the proposed framework. The experimental results demonstrated that the framework enabled reliable vision-language-guided full-body coordination for humanoid robots.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。