arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.03392cs.RO
收藏

OpenEAI-Platform: 一个开源具身人工智能硬件-软件统一平台

OpenEAI-Platform: An Open-source Embodied Artificial Intelligence Hardware-Software Unified Platform

Shanghai Innovation Institute(上海创新研究院);Huazhong University of Science and Technology(华中科技大学);Shanghai Jiao Tong University(上海交通大学)

Jinyuan Zhang, Luoyi Fan, Leiyu Wang, Yeqiang Wang, Yicheng Zhu, Cewu Lu, Nanyang Ye

AI总结:

提出OpenEAI-Platform,集成低成本6+1自由度机械臂和可复现VLA模型,通过开源设计和两阶段训练在真实操作任务中超越商业臂,性能媲美大规模预训练基线。

摘要:

现实世界中的具身AI需要精确的硬件和稳健的视觉-语言-动作(VLA)策略。我们提出OpenEAI-Platform,一个完全开源平台,集成了低成本6+1自由度机械臂(OpenEAI-Arm)和可复现的VLA模型(OpenEAI-VLA)。OpenEAI-Arm提供开源机械设计以实现低制造成本,并采用柔顺控制方法以提高精度。OpenEAI-VLA基于Qwen3-VL-4B,使用扩散Transformer动作头,并仅使用开源机器人和多模态数据集进行两阶段训练。在四个真实操作任务中,OpenEAI-Arm在相同策略下优于两款商用6+1自由度机械臂,而OpenEAI-VLA在仅有限预训练数据下达到了与大规模预训练pi0基线相当的成功率。我们将发布完整的硬件设计、驱动程序、模型以及训练/数据流水线,以支持可复现研究和可扩展数据收集。我们的代码、布局和模型将在论文被接收后发布。

英文摘要:

Embodied AI in the real world requires both accurate hardware and robust vision-language-action (VLA) policies. We present OpenEAI-Platform, a fully open-source platform that integrates a low-cost 6+1 degree-of-freedom (dof) robotic arm (OpenEAI-Arm) and a reproducible VLA model (OpenEAI-VLA). OpenEAI-Arm provides open-source mechanical designs for low manufacturing cost and compliant control methods for higher accuracy. OpenEAI-VLA builds on Qwen3-VL-4B and uses a Diffusion Transformer action head, and is trained in two stages with only open-source robot and multimodal datasets. Across four real-world manipulation tasks, OpenEAI-Arm outperforms two commercial 6+1-dof arms under the same policy, and OpenEAI-VLA achieves success rates comparable to the large-scale pretrained pi0 baseline with only limited pretraining data. We will release the full hardware designs, drivers, models, and training/data pipelines to support reproducible research and scalable data collection. Our codes, layouts, and models will be released after the paper is accepted.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑