让机器人抓取通用化，适配未见过的机械臂形态。
GeoMatch++: Morphology Conditioned Geometry Matching for Multi-Embodiment Grasping
- 用注意力机制学习机械臂结构与物体几何的关系
- 在3种新末端执行器上抓取成功率平均提升9.64%
- 适合需要多类型机械臂通用抓取的场景
尽管近期在多指灵巧抓取方面取得进展，现有方法仍主要针对单一夹爪和未见物体，即使探索跨形态抓取的模型也难以泛化到未见过的末端执行器。本文通过统一策略，学习夹爪形态与物体几何之间的关联，以实现对未见末端执行器的灵巧抓取泛化。机器人形态包含关节与连杆间连接与运动关系的丰富信息，我们利用注意力机制提取更优的末端执行器几何特征。实验表明，相较于先前方法，在3种域外末端执行器上的抓取成功率平均提升9.64%。
原文摘要 · Abstract (English)
Despite recent progress on multi-finger dexterous grasping, current methods focus on single grippers and unseen objects, and even the ones that explore cross-embodiment, often fail to generalize well to unseen end-effectors. This work addresses the problem of dexterous grasping generalization to unseen end-effectors via a unified policy that learns correlation between gripper morphology and object geometry. Robot morphology contains rich information representing how joints and links connect and move with respect to each other and thus, we leverage it through attention to learn better end-effector geometry features. Our experiments show an average of 9.64% increase in grasp success rate across 3 out-of-domain end-effectors compared to previous methods.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。