发表机构
ETH Zurich（苏黎世联邦理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出一种端到端自主流水线，结合生成式AI与可微几何修复，实现按需机器人装配，显著提升装配成功率并支持物理构建。
AI 中文摘要
从数字设计过渡到机器人装配过程目前需要数月专家手动调校，以协调部件几何形状与机器人约束。本文提出了一种端到端的自主流水线，用于定制木质装配体的设计与物理构建。一个生成式AI智能体将用户提示转换为初始3D几何形状，在设计的视觉保真度与特定物理约束之间取得平衡。设计的可装配性通过一个基于梯度的修复阶段进一步改进，该阶段通过图注意力网络代理反向传播以调整部件几何形状。除了修正分离和重叠的部件外，我们展示了硬件特定的修正，可微优化部件几何形状以实现机器人螺丝驱动，在60个新颖自然语言输入中成功率达86.7%，显著优于先前工作十倍。对于其中十个结构，我们使用两台UR5e机器人物理演示了可装配性。这项工作标志着向按需机器人制造迈出的重要一步，能够快速生产定制化、小批量的商品。
英文摘要
Transitioning from a digital design to a robotic assembly process currently requires months of expert manual tuning to reconcile part geometries with robotic constraints. This paper presents an end-to-end, autonomous pipeline for the design and physical construction of bespoke wooden assemblies. A generative AI agent translates user prompts into initial 3D geometries, balancing the visual fidelity of the design with select physical constraints. The assemblability of the design is further improved by a gradient-based repair stage that backpropagates through a graph attention network surrogate to adjust component geometries. In addition to correcting for disjointed and overlapping components, we demonstrate hardware-specific corrections, differentiably optimizing the geometry of components to enable robot screwdriving for 86.7% of 60 novel natural language inputs, significantly outperforming prior work by a factor of ten. For ten of the structures, we physically demonstrate assemblability with two UR5e robots. This work marks a meaningful step toward on-demand robotic manufacturing, enabling the rapid production of customized, low-volume goods.
Comments8 pages, 4 figures, and 2 tables
