[论文] Pixel: A Co-Designed Full-Stack Robot System for Learning General-Purp...

论文概要

研究领域: CV 作者: Tony Tao, Jason Jingzhou Liu, Ruslan Salakhutdinov, Deepak Pathak 发布时间: 2026-11-02 arXiv: 2611.03648

中文摘要

通用移动操作需要机器人硬件、可扩展的示范采集、表达力强的全身控制与有效的策略学习协同工作，但这些组件往往被独立开发，组件间的不匹配限制了系统整体能力。我们提出 Pixel——一套为学习通用移动操作而协同设计的全栈系统。Pixel 集成了低成本轮式人形本体（人形对齐运动学、柔顺夹爪）、多种遥操作与无机器人的人类数据采集接口、统一的全身控制器，以及一套同时适用于人类与机器人示范的策略学习管线。分层学习栈进一步支持语言条件的视觉运动控制与长程任务执行。我们从硬件、数据采集与策略学习三个维度对系统进行评估，展示了需要灵巧操作、主动感知、导航与全身协调的移动操作行为。我们希望 Pixel 为研究机器人硬件、数据、控制与学习如何一体化设计提供一个可扩展的基础。更多结果见 https://pixel-robot.com/pixel

原文摘要

General-purpose mobile manipulation requires a combination of capable robot hardware, scalable demonstration collection, expressive whole-body control, and effective policy learning. Yet these components are often developed independently, creating mismatches that limit the capabilities of the complete system. We present Pixel, a co-designed full-stack system for learning general-purpose mobile manipulation. Pixel combines a low-cost wheeled humanoid with human-aligned kinematics and compliant grippers, multiple teleoperation and robot-free human data-collection interfaces, a unified whole-body controller, and a common policy learning pipeline across human and robot demonstrations. A hierarchical learning stack further enables language-conditioned visuomotor control and long-horizon task ex...

#论文 #arXiv #CV #小凯