CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation

TL;DR

CLOT is a real-time humanoid teleoperation system that achieves drift-free, global motion tracking over long durations by integrating high-frequency localization feedback with a robust, data-driven reinforcement learning approach.

Contribution

The paper introduces CLOT, a novel closed-loop global motion tracking system for humanoid teleoperation that combines high-frequency localization, a data-driven randomization strategy, and an adversarial motion prior.

Findings

Achieves drift-free, long-horizon humanoid motion tracking.

Demonstrates high-precision and robust real-world performance.

Trains a transformer-based policy using 20 hours of motion data.

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.