机器人与具身智能 颠覆级 有讲解视频
发表时间
2026-07-28
arXiv
2607.25895

收录解读

机器人操作策略依赖高质量示范,但真实机器人遥操作昂贵,廉价的无机器人UMI数据又通常只能用于预训练,部署前仍需真实机器人数据锚定。

HiFi-UMI协同设计硬件与软件,使用离线双目惯性SLAM、原生双手相对位姿、微秒级GPIO同步和每只手双广角相机,将无机器人采集精度提升到可直接部署水平。

系统达到约3毫米末端精度;只用HiFi-UMI后训练的策略在三种VLA/WAM骨干上可直接部署,和场内遥操作基线相差-2.5、+3.1和-0.6个百分点,最强插入任务成功率85%。4000小时预训练在10个未见任务上把动作误差降低41%,并开放2000小时数据。

它把无机器人数据从预训练辅助提升为可承担部署后训练的数据来源,形成可复用的数据生产范式。限制在机器人操作领域且工程系统较重,因此未达到paradigm。

原始摘要

Simple AI Learning deployable manipulation policies is bottlenecked by the scarcity of data that is both highfidelity and scalable. Real-robot teleoperation is accurate but costly to scale; robot-free UMI capture scales readily, and current practice uses the resulting data mainly for pre-training, adding a small real-robot “anchor” at post-training time. We ask whether raising the fidelity of robot-free UMI data—rather than shrinking the real-robot fraction—can remove that anchor. We present HiFi-UMI, a portable UMI data-production system that co-designs hardware and software for trajectory accuracy, inter-gripper relative pose, synchronization, and field of view: head-mounted offline stereo-inertial SLAM, native rather than reconstructed relative pose, a shared microsecond GPIO trigger, and two wide-angle cameras per hand covering ∼200◦ . It reaches 3 mm workspace-local end-effector accuracy with no external tracking infrastructure. Using this corpus, we demonstrate zero-robot post-training: a policy post-trained solely on HiFi-UMI demonstrations deploys directly on a real robot and matches in-domain teleoperation across three backbones spanning the vision-language-action (VLA) and worldaction-model (WAM) families, with success-rate differences of −2.5, +3.1, and −0.6 percentage points on StarVLA-QwenPI, OpenPI-π0.5 , and LingBot-VA, respectively; the strongest policy reaches 85% on a precision insertion task—even though the teleoperation baseline is collected in the evaluation scene and no HiFi-UMI trajectory is. Pre-training on 4,000 hours from the same corpus lowers action error on ten unseen tasks by 41% and, on StarVLA-QwenPI, raises real-robot success by a further 18.1 percentage points. We open-source HiFi-UMI-2K—2,000 hours of microsecond-synchronized, ultra-wide-FoV demonstrations, each automatically reconstructed and validated through simulation replay—as a large-scale, high-fidelity resource for the robot-learning community. Website: https://cloud.simpleai.tech/simple-world-lab/hifi-umi/ Dataset: https://huggingface.co/datasets/simple-world-lab/HiFi-UMI-2K

解读视频

相关论文

链接