OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915822085079040 |
|---|---|
| author | Wang, Yunshen Zhu, Shaohang Zhi, Peiyuan Li, Yuhan Li, Jiaxin Li, Yong-Lu Xiao, Yuchen Wang, Xingxing Jia, Baoxiong Huang, Siyuan |
| author_facet | Wang, Yunshen Zhu, Shaohang Zhi, Peiyuan Li, Yuhan Li, Jiaxin Li, Yong-Lu Xiao, Yuchen Wang, Xingxing Jia, Baoxiong Huang, Siyuan |
| contents | High-fidelity motion tracking serves as the ultimate litmus test for generalizable, human-level motor skills. However, current policies often hit a "generality barrier": as motion libraries scale in diversity, tracking fidelity inevitably collapses - especially for real-world deployment of high-dynamic motions. We identify this failure as the result of two compounding factors: the learning bottleneck in scaling multi-motion optimization and the physical executability constraints that arise in real-world actuation. To overcome these challenges, we introduce OmniXtreme, a scalable framework that decouples general motor skill learning from sim-to-real physical skill refinement. Our approach uses a flow-matching policy with high-capacity architectures to scale representation capacity without interference-intensive multi-motion RL optimization, followed by an actuation-aware refinement phase that ensures robust performance on physical hardware. Extensive experiments demonstrate that OmniXtreme maintains high-fidelity tracking across diverse, high-difficulty datasets. On real robots, the unified policy successfully executes multiple extreme motions, effectively breaking the long-standing fidelity-scalability trade-off in high-dynamic humanoid control. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_23843 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control Wang, Yunshen Zhu, Shaohang Zhi, Peiyuan Li, Yuhan Li, Jiaxin Li, Yong-Lu Xiao, Yuchen Wang, Xingxing Jia, Baoxiong Huang, Siyuan Robotics High-fidelity motion tracking serves as the ultimate litmus test for generalizable, human-level motor skills. However, current policies often hit a "generality barrier": as motion libraries scale in diversity, tracking fidelity inevitably collapses - especially for real-world deployment of high-dynamic motions. We identify this failure as the result of two compounding factors: the learning bottleneck in scaling multi-motion optimization and the physical executability constraints that arise in real-world actuation. To overcome these challenges, we introduce OmniXtreme, a scalable framework that decouples general motor skill learning from sim-to-real physical skill refinement. Our approach uses a flow-matching policy with high-capacity architectures to scale representation capacity without interference-intensive multi-motion RL optimization, followed by an actuation-aware refinement phase that ensures robust performance on physical hardware. Extensive experiments demonstrate that OmniXtreme maintains high-fidelity tracking across diverse, high-difficulty datasets. On real robots, the unified policy successfully executes multiple extreme motions, effectively breaking the long-standing fidelity-scalability trade-off in high-dynamic humanoid control. |
| title | OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control |
| topic | Robotics |
| url | https://arxiv.org/abs/2602.23843 |