arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.09016cs.RO
收藏

基于生成式期望的机器人操作闭环视觉运动控制

Closed-Loop Visuomotor Control with Generative Expectation for Robotic Manipulation

Shanghai AI Lab(上海人工智能实验室);Shanghai Jiao Tong University(上海交通大学);HKU(香港大学);Tsinghua University(清华大学)

Qingwen Bu, Jia Zeng, Li Chen, Yanchao Yang, Guyue Zhou, Junchi Yan, Ping Luo, Heming Cui, Yi Ma, Hongyang Li

AI总结:

针对长程任务开环控制易累积误差的问题,提出闭环视觉运动控制框架CLOVER,结合视频扩散模型生成视觉计划、可测嵌入空间量化误差及反馈驱动控制器优化动作,在CALVIN基准上较开环方法提升8%。

摘要:

尽管近年来机器人和具身AI取得了显著进展,但在长程任务中部署机器人仍然是一个巨大挑战。大多数先前的工作遵循开环理念,缺乏实时反馈,导致误差累积和鲁棒性不佳。少数方法尝试利用像素级差异或预训练视觉表示建立反馈机制,但其有效性和适应性被发现受到限制。受经典闭环控制系统启发,我们提出了CLOVER,一个闭环视觉运动控制框架,该框架结合反馈机制以改进自适应机器人控制。CLOVER由一个文本条件视频扩散模型(用于生成视觉计划作为参考输入)、一个可测量的嵌入空间(用于准确量化误差)以及一个反馈驱动控制器(从反馈中优化动作并在需要时启动重新规划)组成。我们的框架在真实世界机器人任务中展现出显著进步,并在CALVIN基准上达到SOTA,较先前的开环方法提升了8%。代码和检查点维护于 https://github.com/OpenDriveLab/CLOVER。

英文摘要:

Despite significant progress in robotics and embodied AI in recent years, deploying robots for long-horizon tasks remains a great challenge. Majority of prior arts adhere to an open-loop philosophy and lack real-time feedback, leading to error accumulation and undesirable robustness. A handful of approaches have endeavored to establish feedback mechanisms leveraging pixel-level differences or pre-trained visual representations, yet their efficacy and adaptability have been found to be constrained. Inspired by classic closed-loop control systems, we propose CLOVER, a closed-loop visuomotor control framework that incorporates feedback mechanisms to improve adaptive robotic control. CLOVER consists of a text-conditioned video diffusion model for generating visual plans as reference inputs, a measurable embedding space for accurate error quantification, and a feedback-driven controller that refines actions from feedback and initiates replans as needed. Our framework exhibits notable advancement in real-world robotic tasks and achieves state-of-the-art on CALVIN benchmark, improving by 8% over previous open-loop counterparts. Code and checkpoints are maintained at https://github.com/OpenDriveLab/CLOVER.

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。

↑