Advancing Humanoid Locomotion: Mastering Challenging Terrains with Denoising World Model Learning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909296042704896 |
|---|---|
| author | Gu, Xinyang Wang, Yen-Jen Zhu, Xiang Shi, Chengming Guo, Yanjiang Liu, Yichen Chen, Jianyu |
| author_facet | Gu, Xinyang Wang, Yen-Jen Zhu, Xiang Shi, Chengming Guo, Yanjiang Liu, Yichen Chen, Jianyu |
| contents | Humanoid robots, with their human-like skeletal structure, are especially suited for tasks in human-centric environments. However, this structure is accompanied by additional challenges in locomotion controller design, especially in complex real-world environments. As a result, existing humanoid robots are limited to relatively simple terrains, either with model-based control or model-free reinforcement learning. In this work, we introduce Denoising World Model Learning (DWL), an end-to-end reinforcement learning framework for humanoid locomotion control, which demonstrates the world's first humanoid robot to master real-world challenging terrains such as snowy and inclined land in the wild, up and down stairs, and extremely uneven terrains. All scenarios run the same learned neural network with zero-shot sim-to-real transfer, indicating the superior robustness and generalization capability of the proposed method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_14472 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Advancing Humanoid Locomotion: Mastering Challenging Terrains with Denoising World Model Learning Gu, Xinyang Wang, Yen-Jen Zhu, Xiang Shi, Chengming Guo, Yanjiang Liu, Yichen Chen, Jianyu Robotics Artificial Intelligence Systems and Control Humanoid robots, with their human-like skeletal structure, are especially suited for tasks in human-centric environments. However, this structure is accompanied by additional challenges in locomotion controller design, especially in complex real-world environments. As a result, existing humanoid robots are limited to relatively simple terrains, either with model-based control or model-free reinforcement learning. In this work, we introduce Denoising World Model Learning (DWL), an end-to-end reinforcement learning framework for humanoid locomotion control, which demonstrates the world's first humanoid robot to master real-world challenging terrains such as snowy and inclined land in the wild, up and down stairs, and extremely uneven terrains. All scenarios run the same learned neural network with zero-shot sim-to-real transfer, indicating the superior robustness and generalization capability of the proposed method. |
| title | Advancing Humanoid Locomotion: Mastering Challenging Terrains with Denoising World Model Learning |
| topic | Robotics Artificial Intelligence Systems and Control |
| url | https://arxiv.org/abs/2408.14472 |