Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2509.05581 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912574256185344 |
|---|---|
| author | Alvarez, Arturo Flores Zargarbashi, Fatemeh Liu, Havel Wang, Shiqi Edwards, Liam Anz, Jessica Xu, Alex Shi, Fan Coros, Stelian Hong, Dennis W. |
| author_facet | Alvarez, Arturo Flores Zargarbashi, Fatemeh Liu, Havel Wang, Shiqi Edwards, Liam Anz, Jessica Xu, Alex Shi, Fan Coros, Stelian Hong, Dennis W. |
| contents | We present a Reinforcement Learning (RL)-based locomotion system for Cosmo, a custom-built humanoid robot designed for entertainment applications. Unlike traditional humanoids, entertainment robots present unique challenges due to aesthetic-driven design choices. Cosmo embodies these with a disproportionately large head (16% of total mass), limited sensing, and protective shells that considerably restrict movement. To address these challenges, we apply Adversarial Motion Priors (AMP) to enable the robot to learn natural-looking movements while maintaining physical stability. We develop tailored domain randomization techniques and specialized reward structures to ensure safe sim-to-real, protecting valuable hardware components during deployment. Our experiments demonstrate that AMP generates stable standing and walking behaviors despite Cosmo's extreme mass distribution and movement constraints. These results establish a promising direction for robots that balance aesthetic appeal with functional performance, suggesting that learning-based methods can effectively adapt to aesthetic-driven design constraints. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_05581 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning to Walk in Costume: Adversarial Motion Priors for Aesthetically Constrained Humanoids Alvarez, Arturo Flores Zargarbashi, Fatemeh Liu, Havel Wang, Shiqi Edwards, Liam Anz, Jessica Xu, Alex Shi, Fan Coros, Stelian Hong, Dennis W. Robotics Artificial Intelligence Systems and Control 68T40 I.2.9; I.2.6 We present a Reinforcement Learning (RL)-based locomotion system for Cosmo, a custom-built humanoid robot designed for entertainment applications. Unlike traditional humanoids, entertainment robots present unique challenges due to aesthetic-driven design choices. Cosmo embodies these with a disproportionately large head (16% of total mass), limited sensing, and protective shells that considerably restrict movement. To address these challenges, we apply Adversarial Motion Priors (AMP) to enable the robot to learn natural-looking movements while maintaining physical stability. We develop tailored domain randomization techniques and specialized reward structures to ensure safe sim-to-real, protecting valuable hardware components during deployment. Our experiments demonstrate that AMP generates stable standing and walking behaviors despite Cosmo's extreme mass distribution and movement constraints. These results establish a promising direction for robots that balance aesthetic appeal with functional performance, suggesting that learning-based methods can effectively adapt to aesthetic-driven design constraints. |
| title | Learning to Walk in Costume: Adversarial Motion Priors for Aesthetically Constrained Humanoids |
| topic | Robotics Artificial Intelligence Systems and Control 68T40 I.2.9; I.2.6 |
| url | https://arxiv.org/abs/2509.05581 |