Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.04412 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908887089676288 |
|---|---|
| author | Wang, Puyue Hu, Jiawei Gao, Yan Wang, Junyan Zhang, Yu Dobbie, Gillian Gu, Tao Johal, Wafa Dang, Ting Jia, Hong |
| author_facet | Wang, Puyue Hu, Jiawei Gao, Yan Wang, Junyan Zhang, Yu Dobbie, Gillian Gu, Tao Johal, Wafa Dang, Ting Jia, Hong |
| contents | Humanoid robots can suffer significant performance drops under small changes in dynamics, task specifications, or environment setup. We propose HoRD, a two-stage learning framework for robust humanoid control under domain shift. First, we train a high-performance teacher policy via history-conditioned reinforcement learning, where the policy infers latent dynamics context from recent state--action trajectories to adapt online to diverse randomized dynamics. Second, we perform online distillation to transfer the teacher's robust control capabilities into a transformer-based student policy that operates on sparse root-relative 3D joint keypoint trajectories. By combining history-conditioned adaptation with online distillation, HoRD enables a single policy to adapt zero-shot to unseen domains without per-domain retraining. Extensive experiments show HoRD outperforms strong baselines in robustness and transfer, especially under unseen domains and external perturbations. Code and project page are available at https://tonywang-0517.github.io/hord/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_04412 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation Wang, Puyue Hu, Jiawei Gao, Yan Wang, Junyan Zhang, Yu Dobbie, Gillian Gu, Tao Johal, Wafa Dang, Ting Jia, Hong Robotics Machine Learning Humanoid robots can suffer significant performance drops under small changes in dynamics, task specifications, or environment setup. We propose HoRD, a two-stage learning framework for robust humanoid control under domain shift. First, we train a high-performance teacher policy via history-conditioned reinforcement learning, where the policy infers latent dynamics context from recent state--action trajectories to adapt online to diverse randomized dynamics. Second, we perform online distillation to transfer the teacher's robust control capabilities into a transformer-based student policy that operates on sparse root-relative 3D joint keypoint trajectories. By combining history-conditioned adaptation with online distillation, HoRD enables a single policy to adapt zero-shot to unseen domains without per-domain retraining. Extensive experiments show HoRD outperforms strong baselines in robustness and transfer, especially under unseen domains and external perturbations. Code and project page are available at https://tonywang-0517.github.io/hord/. |
| title | HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2602.04412 |