Learning Humanoid Locomotion with World Model Reconstruction
Wandong Sun, Long Chen, Yongbo Su, Baoshi Cao, Yang Liu, Zongwu Xie · 2025
Read Learning Humanoid Locomotion with World Model Reconstruction by Wandong Sun, Long Chen, Yongbo Su, Baoshi Cao, Yang Liu, Zongwu Xie (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-16230.txt.
Description
Wandong Sun and co-authors taught a humanoid robot to walk over hard terrain in a 2025 paper. Lab controllers often fail on rough real ground, partly because sensor data are noisy and limited. Their method, World Model Reconstruction, trains an estimator to rebuild the world state explicitly.
The walking policy then acts only on that rebuilt information, not on raw sensor data. Policy and estimator train together, but the learning signal between them is cut, so the estimator stays focused on accurate rebuilding. Tests covered rough, soft, and slippery ground. The robot finished a hike of over three kilometres with no human help, crossing ice and snow.
Catalog record
| CREATOR | Wandong Sun, Long Chen, Yongbo Su, Baoshi Cao, Yang Liu, Zongwu Xie |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-16230