MechaFormer：面向运动学机构设计自动化的序列学习
MechaFormer: Sequence Learning for Kinematic Mechanism Design Automation
AI总结：
针对机械机构路径设计搜索空间复杂的难题，提出基于Transformer的MechaFormer模型，将机构设计转化为条件序列生成任务，在统一过程中确定拓扑与参数，显著提升路径匹配精度与设计多样性，并可与传统优化器结合高效求解。
摘要：
设计用于描绘特定路径的机械机构是一个经典但极其困难的工程问题，其特征是离散拓扑和连续参数构成了庞大且复杂的搜索空间。我们引入了 MechaFormer，这是一种基于 Transformer 的模型，它通过将机构设计视为条件序列生成任务来应对这一挑战。我们的模型学习将目标曲线转换为领域特定语言（DSL）字符串，在单一统一的过程中同时确定机构的拓扑结构和几何参数。MechaFormer 显著优于现有基线，实现了最先进的路径匹配精度，并生成了广泛多样的新颖且有效的设计。我们演示了一套采样策略，这些策略能显著提高解的质量，并为设计者提供宝贵的灵活性。此外，我们表明 MechaFormer 的高质量输出可作为传统优化器的极佳起点，创造了一种混合方法，以卓越的效率找到更优的解。
英文摘要：
Designing mechanical mechanisms to trace specific paths is a classic yet notoriously difficult engineering problem, characterized by a vast and complex search space of discrete topologies and continuous parameters. We introduce MechaFormer, a Transformer-based model that tackles this challenge by treating mechanism design as a conditional sequence generation task. Our model learns to translate a target curve into a domain-specific language (DSL) string, simultaneously determining the mechanism's topology and geometric parameters in a single, unified process. MechaFormer significantly outperforms existing baselines, achieving state-of-the-art path-matching accuracy and generating a wide diversity of novel and valid designs. We demonstrate a suite of sampling strategies that can dramatically improve solution quality and offer designers valuable flexibility. Furthermore, we show that the high-quality outputs from MechaFormer serve as excellent starting points for traditional optimizers, creating a hybrid approach that finds superior solutions with remarkable efficiency.
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。
