外观
论文arXiv 2602.150608 个月前
CLOT系统实现全尺寸人形机器人长时遥操作全局无漂移跟踪
本文由 AI 根据公开内容整理，可能有误，请以原文为准：CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation（arXiv 2602.15060）
核心内容
针对全尺寸人形机器人长时遥操作中累积全局位姿漂移导致的不稳定问题，研究团队提出CLOT实时全身遥操作系统。该系统通过高频定位反馈构建闭环，解决了传统学习方法仅依赖局部坐标系导致的长时序执行偏差。为支撑策略训练，团队采集并清洗了20小时高质量人体运动数据，投入超1300 GPU小时完成模型迭代，并在仿真与真机实验中验证了其在高动态动作下的精准跟踪能力与Sim-to-Real鲁棒性。
技术层面，CLOT采用数据驱动随机化策略，将观测轨迹与奖励评估解耦，避免了直接施加全局跟踪奖励导致的激进修正与策略脆弱性；同时引入对抗性运动先验正则化项以抑制非自然行为。系统部署于31自由度（不含手）全尺寸人形机器人，采用Transformer架构策略网络。相关运动数据集、演示视频及源代码已开源，为具身智能遥操作提供了从数据到算法的完整工程参考。
关键看点
- 闭环全局跟踪：利用高频定位反馈同步操作员与机器人位姿，解决长时遥操作中的全局漂移与不稳定问题
- 奖励解耦策略：采用数据驱动随机化解耦观测与奖励，配合对抗性运动先验，实现平滑稳定的全局修正
- 数据与算力规模：构建20小时精选人体运动数据集，Transformer策略训练耗时超1300 GPU小时
- 本体与开源：在31 DoF全尺寸人形机器人验证高动态Sim-to-Real性能，代码与数据已开放获取
工程与产业观察
CLOT证明了在全局坐标系下引入高频定位反馈对解决人形机器人长时遥操作漂移的有效性。其提出的观测-奖励解耦与对抗性先验机制，为强化学习在真实物理系统中的稳定部署提供了关键范式。20小时动捕数据与31 DoF真机验证的组合，显著提升了该方案在具身数据采集与远程操控场景下的工程参考价值。
英文摘要 (Abstract)
Long-horizon whole-body humanoid teleoperation remains challenging due to accumulated global pose drift, particularly on full-sized humanoids. Although recent learning-based tracking methods enable agile and coordinated motions, they typically operate in the robot's local frame and neglect global pose feedback, leading to drift and instability during extended execution. In this work, we present CLOT, a real-time whole-body humanoid teleoperation system that achieves closed-loop global motion tracking via high-frequency localization feedback. CLOT synchronizes operator and robot poses in a closed loop, enabling drift-free human-to-humanoid mimicry over long timehorizons. However, directly imposing global tracking rewards in reinforcement learning, often results in aggressive and brittle corrections. To address this, we propose a data-driven randomization strategy that decouples observation trajectories from reward evaluation, enabling smooth and stable global corrections. We further regularize the policy with an adversarial motion prior to suppress unnatural behaviors. To support CLOT, we collect 20 hours of carefully curated human motion data for training the humanoid teleoperation policy. We design a transformer-based policy and train it for over 1300 GPU hours. The policy is deployed on a full-sized humanoid with 31 DoF (excluding hands). Both simulation and real-world experiments verify high-dynamic motion, high-precision tracking, and strong robustness in sim-to-real humanoid teleoperation. Motion data, demos and code can be found in our website.
作者：Tengjie Zhu, Guanyu Cai, Yang Zhaohui, Guanzhu Ren, Haohui Xie, ZiRui Wang, Junsong Wu, Jingbo Wang, Xiaokang Yang, Yao Mu, Yichao Yan
来源
本页事实对应的公开页面，以原文为准。