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 2409.15517cs.ROcs.CV
收藏

MATCH POLICY:从点云配准到操作策略的简单流水线

MATCH POLICY: A Simple Pipeline from Point Cloud Registration to Manipulation Policies

Khoury College of Computer Science, Northeastern University(东北大学科里计算机科学学院);Worcester Polytechnic Institute(伍斯特理工学院)

Haojie Huang, Haotian Liu, Dian Wang, Robin Walters, Robert Platt

AI总结:

提出MATCH POLICY流水线,将抓取与放置目标配准到存储演示中,把动作推断转化为点云配准任务,无需训练即可实现高精度操作策略,在RLBench基准和真实机器人上均达到最先进性能。

摘要:

许多操作任务要求机器人对物体进行相对重排。此类任务可以描述为一组刚体各部分之间的一系列相对位姿。在本研究中,我们提出了MATCH POLICY,一种简单但新颖的流水线,用于解决高精度的抓取与放置任务。我们的方法不直接预测动作,而是将抓取与放置目标配准到存储的演示中。这将动作推断转化为点云配准任务,并使我们能够在无需任何训练的情况下实现非平凡的操作策略。MATCH POLICY旨在通过关键帧设置解决高精度任务。通过利用几何交互和任务的对称性,它实现了极高的样本效率,并对未见过的配置具有泛化能力。我们在RLBench基准上与多个强基线相比展示了其在各种任务上的最先进性能,并在真实机器人上用六个任务对其进行了测试。

英文摘要:

Many manipulation tasks require the robot to rearrange objects relative to one another. Such tasks can be described as a sequence of relative poses between parts of a set of rigid bodies. In this work, we propose MATCH POLICY, a simple but novel pipeline for solving high-precision pick and place tasks. Instead of predicting actions directly, our method registers the pick and place targets to the stored demonstrations. This transfers action inference into a point cloud registration task and enables us to realize nontrivial manipulation policies without any training. MATCH POLICY is designed to solve high-precision tasks with a key-frame setting. By leveraging the geometric interaction and the symmetries of the task, it achieves extremely high sample efficiency and generalizability to unseen configurations. We demonstrate its state-of-the-art performance across various tasks on RLBench benchmark compared with several strong baselines and test it on a real robot with six tasks.

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。

↑