Learning Quadruped Walking from Seconds of Demonstration
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914377853042688 |
|---|---|
| author | Zhang, Ruipeng Yu, Hongzhan Chang, Ya-Chien Li, Chenghao Christensen, Henrik I. Gao, Sicun |
| author_facet | Zhang, Ruipeng Yu, Hongzhan Chang, Ya-Chien Li, Chenghao Christensen, Henrik I. Gao, Sicun |
| contents | Quadruped locomotion provides a natural setting for understanding when model-free learning can outperform model-based control design, by exploiting data patterns to bypass the difficulty of optimizing over discrete contacts and the combinatorial explosion of mode changes. We give a principled analysis of why imitation learning with quadrupeds can be inherently effective in a small data regime, based on the structure of its limit cycles, Poincaré return maps, and local numerical properties of neural networks. The understanding motivates a new imitation learning method that regulates the alignment between variations in a latent space and those over the output actions. Hardware experiments confirm that a few seconds of demonstration is sufficient to train various locomotion policies from scratch entirely offline with reasonable robustness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_06961 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Quadruped Walking from Seconds of Demonstration Zhang, Ruipeng Yu, Hongzhan Chang, Ya-Chien Li, Chenghao Christensen, Henrik I. Gao, Sicun Machine Learning Artificial Intelligence Quadruped locomotion provides a natural setting for understanding when model-free learning can outperform model-based control design, by exploiting data patterns to bypass the difficulty of optimizing over discrete contacts and the combinatorial explosion of mode changes. We give a principled analysis of why imitation learning with quadrupeds can be inherently effective in a small data regime, based on the structure of its limit cycles, Poincaré return maps, and local numerical properties of neural networks. The understanding motivates a new imitation learning method that regulates the alignment between variations in a latent space and those over the output actions. Hardware experiments confirm that a few seconds of demonstration is sufficient to train various locomotion policies from scratch entirely offline with reasonable robustness. |
| title | Learning Quadruped Walking from Seconds of Demonstration |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.06961 |