SMAP: Self-supervised Motion Adaptation for Physically Plausible Humanoid Whole-body Control
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910968536104960 |
|---|---|
| author | Zhao, Haoyu Lin, Sixu Ben, Qingwei Dai, Minyue Fei, Hao Wang, Jingbo Zou, Hua Dong, Junting |
| author_facet | Zhao, Haoyu Lin, Sixu Ben, Qingwei Dai, Minyue Fei, Hao Wang, Jingbo Zou, Hua Dong, Junting |
| contents | This paper presents a novel framework that enables real-world humanoid robots to maintain stability while performing human-like motion. Current methods train a policy which allows humanoid robots to follow human body using the massive retargeted human data via reinforcement learning. However, due to the heterogeneity between human and humanoid robot motion, directly using retargeted human motion reduces training efficiency and stability. To this end, we introduce SMAP, a novel whole-body tracking framework that bridges the gap between human and humanoid action spaces, enabling accurate motion mimicry by humanoid robots. The core idea is to use a vector-quantized periodic autoencoder to capture generic atomic behaviors and adapt human motion into physically plausible humanoid motion. This adaptation accelerates training convergence and improves stability when handling novel or challenging motions. We then employ a privileged teacher to distill precise mimicry skills into the student policy with a proposed decoupled reward. We conduct experiments in simulation and real world to demonstrate the superiority stability and performance of SMAP over SOTA methods, offering practical guidelines for advancing whole-body control in humanoid robots. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_19463 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SMAP: Self-supervised Motion Adaptation for Physically Plausible Humanoid Whole-body Control Zhao, Haoyu Lin, Sixu Ben, Qingwei Dai, Minyue Fei, Hao Wang, Jingbo Zou, Hua Dong, Junting Robotics This paper presents a novel framework that enables real-world humanoid robots to maintain stability while performing human-like motion. Current methods train a policy which allows humanoid robots to follow human body using the massive retargeted human data via reinforcement learning. However, due to the heterogeneity between human and humanoid robot motion, directly using retargeted human motion reduces training efficiency and stability. To this end, we introduce SMAP, a novel whole-body tracking framework that bridges the gap between human and humanoid action spaces, enabling accurate motion mimicry by humanoid robots. The core idea is to use a vector-quantized periodic autoencoder to capture generic atomic behaviors and adapt human motion into physically plausible humanoid motion. This adaptation accelerates training convergence and improves stability when handling novel or challenging motions. We then employ a privileged teacher to distill precise mimicry skills into the student policy with a proposed decoupled reward. We conduct experiments in simulation and real world to demonstrate the superiority stability and performance of SMAP over SOTA methods, offering practical guidelines for advancing whole-body control in humanoid robots. |
| title | SMAP: Self-supervised Motion Adaptation for Physically Plausible Humanoid Whole-body Control |
| topic | Robotics |
| url | https://arxiv.org/abs/2505.19463 |