HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911519343640576 |
|---|---|
| author | Wang, Xingyi Zhang, Chenyun Xie, Weiji Yu, Chao Song, Wei Bai, Chenjia Zhu, Shiqiang |
| author_facet | Wang, Xingyi Zhang, Chenyun Xie, Weiji Yu, Chao Song, Wei Bai, Chenjia Zhu, Shiqiang |
| contents | Humanoid robots deployed in real-world scenarios often need to carry unknown payloads, which introduce significant mismatch and degrade the effectiveness of simulation-to-reality reinforcement learning methods. To address this challenge, we propose a two-stage gradient-based system identification framework built on the differentiable simulator MuJoCo XLA. The first stage calibrates the nominal robot model using real-world data to reduce intrinsic sim-to-real discrepancies, while the second stage further identifies the mass distribution of the unknown payload. By explicitly reducing structured model bias prior to policy training, our approach enables zero-shot transfer of reinforcement learning policies to hardware under heavy-load conditions. Extensive simulation and real-world experiments demonstrate more precise parameter identification, improved motion tracking accuracy, and substantially enhanced agility and robustness compared to existing baselines. Project Page: https://mwondering.github.io/halo-humanoid/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15084 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulation Wang, Xingyi Zhang, Chenyun Xie, Weiji Yu, Chao Song, Wei Bai, Chenjia Zhu, Shiqiang Robotics Humanoid robots deployed in real-world scenarios often need to carry unknown payloads, which introduce significant mismatch and degrade the effectiveness of simulation-to-reality reinforcement learning methods. To address this challenge, we propose a two-stage gradient-based system identification framework built on the differentiable simulator MuJoCo XLA. The first stage calibrates the nominal robot model using real-world data to reduce intrinsic sim-to-real discrepancies, while the second stage further identifies the mass distribution of the unknown payload. By explicitly reducing structured model bias prior to policy training, our approach enables zero-shot transfer of reinforcement learning policies to hardware under heavy-load conditions. Extensive simulation and real-world experiments demonstrate more precise parameter identification, improved motion tracking accuracy, and substantially enhanced agility and robustness compared to existing baselines. Project Page: https://mwondering.github.io/halo-humanoid/ |
| title | HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.15084 |