GeRM: A Generalist Robotic Model with Mixture-of-experts for Quadruped Robot
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916198216630272 |
|---|---|
| author | Song, Wenxuan Zhao, Han Ding, Pengxiang Cui, Can Lyu, Shangke Fan, Yaning Wang, Donglin |
| author_facet | Song, Wenxuan Zhao, Han Ding, Pengxiang Cui, Can Lyu, Shangke Fan, Yaning Wang, Donglin |
| contents | Multi-task robot learning holds significant importance in tackling diverse and complex scenarios. However, current approaches are hindered by performance issues and difficulties in collecting training datasets. In this paper, we propose GeRM (Generalist Robotic Model). We utilize offline reinforcement learning to optimize data utilization strategies to learn from both demonstrations and sub-optimal data, thus surpassing the limitations of human demonstrations. Thereafter, we employ a transformer-based VLA network to process multi-modal inputs and output actions. By introducing the Mixture-of-Experts structure, GeRM allows faster inference speed with higher whole model capacity, and thus resolves the issue of limited RL parameters, enhancing model performance in multi-task learning while controlling computational costs. Through a series of experiments, we demonstrate that GeRM outperforms other methods across all tasks, while also validating its efficiency in both training and inference processes. Additionally, we uncover its potential to acquire emergent skills. Additionally, we contribute the QUARD-Auto dataset, collected automatically to support our training approach and foster advancements in multi-task quadruped robot learning. This work presents a new paradigm for reducing the cost of collecting robot data and driving progress in the multi-task learning community. You can reach our project and video through the link: https://songwxuan.github.io/GeRM/ . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_13358 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GeRM: A Generalist Robotic Model with Mixture-of-experts for Quadruped Robot Song, Wenxuan Zhao, Han Ding, Pengxiang Cui, Can Lyu, Shangke Fan, Yaning Wang, Donglin Robotics Computer Vision and Pattern Recognition Machine Learning Multi-task robot learning holds significant importance in tackling diverse and complex scenarios. However, current approaches are hindered by performance issues and difficulties in collecting training datasets. In this paper, we propose GeRM (Generalist Robotic Model). We utilize offline reinforcement learning to optimize data utilization strategies to learn from both demonstrations and sub-optimal data, thus surpassing the limitations of human demonstrations. Thereafter, we employ a transformer-based VLA network to process multi-modal inputs and output actions. By introducing the Mixture-of-Experts structure, GeRM allows faster inference speed with higher whole model capacity, and thus resolves the issue of limited RL parameters, enhancing model performance in multi-task learning while controlling computational costs. Through a series of experiments, we demonstrate that GeRM outperforms other methods across all tasks, while also validating its efficiency in both training and inference processes. Additionally, we uncover its potential to acquire emergent skills. Additionally, we contribute the QUARD-Auto dataset, collected automatically to support our training approach and foster advancements in multi-task quadruped robot learning. This work presents a new paradigm for reducing the cost of collecting robot data and driving progress in the multi-task learning community. You can reach our project and video through the link: https://songwxuan.github.io/GeRM/ . |
| title | GeRM: A Generalist Robotic Model with Mixture-of-experts for Quadruped Robot |
| topic | Robotics Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2403.13358 |