AI Agents 相关度: 7/10

SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation

Milo Carroll, Tianhu Peng, Lingfan Bao, Chengxu Zhou, Zhibin Li
arXiv: 2603.09574v1 发布: 2026-03-10 更新: 2026-03-10

AI 摘要

SCDP通过混合观测蒸馏,仅用板载传感器实现了鲁棒的人形机器人运动控制。

主要贡献

  • 提出Sensor-Conditioned Diffusion Policies (SCDP)
  • 混合观测训练:传感器历史条件下的diffusion模型预测未来状态-动作轨迹
  • 引入restricted denoising, context distribution alignment, context-aware attention masking

方法论

使用sensor history作为条件,通过diffusion模型预测未来状态-动作轨迹,并采用多种方法提升模型性能。

原文摘要

Distilling humanoid locomotion control from offline datasets into deployable policies remains a challenge, as existing methods rely on privileged full-body states that require complex and often unreliable state estimation. We present Sensor-Conditioned Diffusion Policies (SCDP) that enables humanoid locomotion using only onboard sensors, eliminating the need for explicit state estimation. SCDP decouples sensing from supervision through mixed-observation training: diffusion model conditions on sensor histories while being supervised to predict privileged future state-action trajectories, enforcing the model to infer the motion dynamics under partial observability. We further develop restricted denoising, context distribution alignment, and context-aware attention masking to encourage implicit state estimation within the model and to prevent train-deploy mismatch. We validate SCDP on velocity-commanded locomotion and motion reference tracking tasks. In simulation, SCDP achieves near-perfect success on velocity control (99-100%) and 93% tracking success in AMASS test set, performing comparable to privileged baselines while using only onboard sensors. Finally, we deploy the trained policy on a real G1 humanoid at 50 Hz, demonstrating robust real robot locomotion without external sensing or state estimation.

标签

人形机器人 运动控制 强化学习 Diffusion模型 部分观测

arXiv 分类

cs.RO cs.LG