arXiv Robotics — research abstracts· Hongzhi Zang, Shu'ang Yu, Hao Lin, Tianxing Zhou, Zefang Huang, Zhen Guo, Xin Xu, Jiakai Zhou, Yuze Sheng, Shizhe Zhang, Feng Gao, Wenhao Tang, Yufeng Yue, Quanlu Zhang, Xinlei Chen, Chao Yu, Yu Wang·· 7 days agoEditorial score63
USER: A Unified and Extensible System for Real-World Online Policy Learning in Embodied AI
USER: A Unified and Extensible System for Real-World Online Policy Learning in Embodied AI
Summary
USER is a system designed for real-world online policy learning in embodied AI, addressing the challenges of real-world environments by integrating hardware abstraction, adaptive communication, and asynchronous training. It supports multi-robot coordination, heterogeneous manipulators, and cloud-edge training with large models, establishing a foundation for real-world policy learning.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org