UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms
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_ | 1866916063764021248 |
|---|---|
| author | Jia, Yufei Cao, Zhanxiang Yu, Mingrui Zhang, Heng Chen, Shenyu Jiang, Dixuan Li, Meng Li, Xiaofan Liu, Yiyang Wu, Junzhe Li, Zheng Fang, XiLin Cui, Tingyu Fu, Shengcheng Li, Haoyang Wang, Anqi Wang, Zifan Zhu, Dongjie Cao, Chenyu Huang, Zhenbiao Zheng, Ziang Lu, Jie Ma, Xin Wei, Zhengyang Zhao, Xiang Zhan, Tianyue He, Ye Chen, Yuxiang Jiang, Yizhou Li, Yue Ge, Haizhou Dong, Yuhang Jia, Fan Zhang, Ziheng Zhang, Meng Deng, Xiwa Chen, Zhixing Shao, Hanyang Dong, Chenxin Li, Yixuan Chen, Yizhi Chen, Bokui Zhang, Kaifeng Cui, Hanqing Qin, Yusen Huang, Ruqi Han, Lei Wang, Tiancai Li, Xiang Gao, Yue Zhou, Guyue |
| author_facet | Jia, Yufei Cao, Zhanxiang Yu, Mingrui Zhang, Heng Chen, Shenyu Jiang, Dixuan Li, Meng Li, Xiaofan Liu, Yiyang Wu, Junzhe Li, Zheng Fang, XiLin Cui, Tingyu Fu, Shengcheng Li, Haoyang Wang, Anqi Wang, Zifan Zhu, Dongjie Cao, Chenyu Huang, Zhenbiao Zheng, Ziang Lu, Jie Ma, Xin Wei, Zhengyang Zhao, Xiang Zhan, Tianyue He, Ye Chen, Yuxiang Jiang, Yizhou Li, Yue Ge, Haizhou Dong, Yuhang Jia, Fan Zhang, Ziheng Zhang, Meng Deng, Xiwa Chen, Zhixing Shao, Hanyang Dong, Chenxin Li, Yixuan Chen, Yizhi Chen, Bokui Zhang, Kaifeng Cui, Hanqing Qin, Yusen Huang, Ruqi Han, Lei Wang, Tiancai Li, Xiang Gao, Yue Zhou, Guyue |
| contents | Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-centric execution path. This paradigm has greatly improved training speed, but it has also encouraged a default assumption that efficient training requires physics to reside on the GPU. We revisit this assumption. Our view is that, in simulation-dominated robot control, the essential question is not which processor runs physics, but whether simulation throughput, policy learning, and runtime synchronization form an efficient end-to-end loop. We present UniLab, a heterogeneous CPU-simulation / GPU-learning architecture that decouples CPU-parallel simulation from GPU policy updates through a unified runtime for data movement, buffering, and synchronization. UniLab is implemented as a complete and extensible training system using MuJoCoUni and MotrixSim CPU-batched physics backends, supporting PPO, FastSAC, FlashSAC, and APPO. On representative simulation-based robot control tasks, UniLab improves end-to-end training efficiency by 3--10$\times$ under the same hardware configuration, while reducing dependence on the NVIDIA CUDA-based software stack and supporting cross-platform execution on the Apple macOS platform and the AMD ROCm and Intel XPU accelerator backends. These results show that GPU simulation is an effective path to efficient training, but not a necessary one, broadening the practical system choices available for robot RL training. Project page: https://unilabsim.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_30313 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms Jia, Yufei Cao, Zhanxiang Yu, Mingrui Zhang, Heng Chen, Shenyu Jiang, Dixuan Li, Meng Li, Xiaofan Liu, Yiyang Wu, Junzhe Li, Zheng Fang, XiLin Cui, Tingyu Fu, Shengcheng Li, Haoyang Wang, Anqi Wang, Zifan Zhu, Dongjie Cao, Chenyu Huang, Zhenbiao Zheng, Ziang Lu, Jie Ma, Xin Wei, Zhengyang Zhao, Xiang Zhan, Tianyue He, Ye Chen, Yuxiang Jiang, Yizhou Li, Yue Ge, Haizhou Dong, Yuhang Jia, Fan Zhang, Ziheng Zhang, Meng Deng, Xiwa Chen, Zhixing Shao, Hanyang Dong, Chenxin Li, Yixuan Chen, Yizhi Chen, Bokui Zhang, Kaifeng Cui, Hanqing Qin, Yusen Huang, Ruqi Han, Lei Wang, Tiancai Li, Xiang Gao, Yue Zhou, Guyue Robotics 68T40 I.2.9 Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-centric execution path. This paradigm has greatly improved training speed, but it has also encouraged a default assumption that efficient training requires physics to reside on the GPU. We revisit this assumption. Our view is that, in simulation-dominated robot control, the essential question is not which processor runs physics, but whether simulation throughput, policy learning, and runtime synchronization form an efficient end-to-end loop. We present UniLab, a heterogeneous CPU-simulation / GPU-learning architecture that decouples CPU-parallel simulation from GPU policy updates through a unified runtime for data movement, buffering, and synchronization. UniLab is implemented as a complete and extensible training system using MuJoCoUni and MotrixSim CPU-batched physics backends, supporting PPO, FastSAC, FlashSAC, and APPO. On representative simulation-based robot control tasks, UniLab improves end-to-end training efficiency by 3--10$\times$ under the same hardware configuration, while reducing dependence on the NVIDIA CUDA-based software stack and supporting cross-platform execution on the Apple macOS platform and the AMD ROCm and Intel XPU accelerator backends. These results show that GPU simulation is an effective path to efficient training, but not a necessary one, broadening the practical system choices available for robot RL training. Project page: https://unilabsim.github.io. |
| title | UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms |
| topic | Robotics 68T40 I.2.9 |
| url | https://arxiv.org/abs/2605.30313 |