Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer
Xue, Haoru, He, Tairan, Wang, Zi, Ben, Qingwei, Xiao, Wenli, Luo, Zhengyi, Da, Xingye, Castañeda, Fernando 等 12 位作者
原文摘要
Recent progress in GPU-accelerated, photorealistic simulation has opened a scalable data-generation path for robot learning, where massive physics and visual randomization allow policies to generalize beyond curated environments. Building on these advances, we develop a teacher-student-bootstrap learning framework for vision-based humanoid loco-manipulation, using articulated-object interaction as a representative high-difficulty benchmark. Our approach introduces a staged-reset exploration strategy that stabilizes long-horizon privileged-policy training, and a GRPO-based fine-tuning procedure designed to mitigate partial observability and improve closed-loop consistency in sim-to-real RL. Trained entirely on synthetic simulation data, the resulting policy achieves robust zero-shot performance across diverse articulated objects--including multiple door types--and outperforms human teleoperators by up to 31.7% in task completion time under the same whole-body control stack. This represents the first humanoid sim-to-real policy capable of diverse articulated loco-manipulation from pure RGB perception.
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表