外观
论文arXiv 2610.0977721 天前
通过可微分几何部件修复实现按需机器人装配
针对数字设计到机器人装配需人工调参的问题，提出生成式AI与可微几何修复流水线，并在木质结构和双UR5e实物装配中验证。
Universal Robots UR5e原文链接：On-Demand Robotic Assembly via Differentiable Geometric Part Repair（arXiv 2610.09777 · 发布日期：2026-10-07）
On-Demand Robotic Assembly via Differentiable Geometric Part Repair
针对数字设计到机器人装配需人工调参的问题，提出生成式AI与可微几何修复流水线，并在木质结构和双UR5e实物装配中验证。
核心要点
- 动机是将数字设计转为机器人装配目前需要数月专家手动调参，以协调部件几何与机器人约束。
- 方法以生成式AI代理把用户提示转为初始3D几何，并用基于图注意力网络代理的梯度修复调整组件几何。
- 修复阶段可纠正分离和重叠组件，并进行硬件相关优化，使60个新自然语言输入中86.7%可机器人拧螺丝。
- 实验称该方法较先前工作提升十倍，并对其中10个结构用两台UR5e机器人进行了实物可装配性演示。
验证范围与局限：验证范围为定制木质装配：60个新自然语言输入的设计优化，以及其中10个结构的双UR5e实物装配演示。
英文摘要 (Abstract)
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.
作者：Millicent Schlafly, Fabio Schaub, Diogo Costa Pais, Luca Lelli, Janne Dvorak, Claire Colmont, Sven Marti, Mark Fuge
来源
本页事实对应的公开页面，以原文为准。