arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.02241cs.RO
收藏

WeightedPose:基于加权SVD的可泛化跨姿态估计

WeightedPose: Generalizable Cross-Pose Estimation via Weighted SVD

Xuxin Cheng, Heng Yu, Harry Zhang, Wenxing Deng

AI总结:

针对传统端到端方法难以理解物体位姿关系、适配新配置的问题,提出WeightedPose方法,用加权SVD学习物体间三维几何关系,提升机器人复杂操作能力。

摘要:

我们提出了一种面向人类环境下机器人操作任务的新方法,该方法需要理解一对物体之间的三维几何关联。传统的端到端训练方法将像素观测直接转换为机器人动作,往往无法有效理解复杂的位姿关系,也难以适配新的物体配置。为解决这些问题,我们的方法聚焦于学习三维几何关系,尤其是一个物体的关键部件与另一个物体的关键部件之间的关联方式。我们在独立模型中采用加权SVD(奇异值分解),分析铰接部件与自由漂浮物体的位姿关系。例如,我们的模型能够理解烤箱门与烤箱本体之间,以及千层面烤盘与烤箱之间的空间关系。通过聚焦三维几何关联,我们的策略使机器人能够基于以物体为中心的视角执行复杂操作任务。

英文摘要:

We introduce a new approach for robotic manipulation tasks in human settings that necessitates understanding the 3D geometric connections between a pair of objects. Conventional end-to-end training approaches, which convert pixel observations directly into robot actions, often fail to effectively understand complex pose relationships and do not easily adapt to new object configurations. To overcome these issues, our method focuses on learning the 3D geometric relationships, particularly how critical parts of one object relate to those of another. We employ Weighted SVD in our standalone model to analyze pose relationships both in articulated parts and in free-floating objects. For instance, our model can comprehend the spatial relationship between an oven door and the oven body, as well as between a lasagna plate and the oven. By concentrating on the 3D geometric connections, our strategy empowers robots to carry out intricate manipulation tasks based on object-centric perspectives

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。

↑