Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.14756 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908662876864512 |
|---|---|
| author | Wei, Lai Peng, Xuanbin Qiu, Ri-Zhao Huang, Tianshu Cheng, Xuxin Wang, Xiaolong |
| author_facet | Wei, Lai Peng, Xuanbin Qiu, Ri-Zhao Huang, Tianshu Cheng, Xuxin Wang, Xiaolong |
| contents | Learning from real-world robot demonstrations holds promise for interacting with complex real-world environments. However, the complexity and variability of interaction dynamics often cause purely positional controllers to struggle with contacts or varying payloads. To address this, we propose a Heterogeneous Meta-Control (HMC) framework for Loco-Manipulation that adaptively stitches multiple control modalities: position, impedance, and hybrid force-position. We first introduce an interface, HMC-Controller, for blending actions from different control profiles continuously in the torque space. HMC-Controller facilitates both teleoperation and policy deployment. Then, to learn a robust force-aware policy, we propose HMC-Policy to unify different controllers into a heterogeneous architecture. We adopt a mixture-of-experts style routing to learn from large-scale position-only data and fine-grained force-aware demonstrations. Experiments on a real humanoid robot show over 50% relative improvement vs. baselines on challenging tasks such as compliant table wiping and drawer opening, demonstrating the efficacy of HMC. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_14756 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation Wei, Lai Peng, Xuanbin Qiu, Ri-Zhao Huang, Tianshu Cheng, Xuxin Wang, Xiaolong Robotics Learning from real-world robot demonstrations holds promise for interacting with complex real-world environments. However, the complexity and variability of interaction dynamics often cause purely positional controllers to struggle with contacts or varying payloads. To address this, we propose a Heterogeneous Meta-Control (HMC) framework for Loco-Manipulation that adaptively stitches multiple control modalities: position, impedance, and hybrid force-position. We first introduce an interface, HMC-Controller, for blending actions from different control profiles continuously in the torque space. HMC-Controller facilitates both teleoperation and policy deployment. Then, to learn a robust force-aware policy, we propose HMC-Policy to unify different controllers into a heterogeneous architecture. We adopt a mixture-of-experts style routing to learn from large-scale position-only data and fine-grained force-aware demonstrations. Experiments on a real humanoid robot show over 50% relative improvement vs. baselines on challenging tasks such as compliant table wiping and drawer opening, demonstrating the efficacy of HMC. |
| title | HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2511.14756 |