[论文解读] LINKS: A dataset of a hundred million planar linkage mechanisms for data-driven kinematic design

Amin Heyrani Nobari, Akash Srivastava+2 more|arXiv (Cornell University)|Aug 30, 2022

Robotic Mechanisms and Dynamics被引用 4

一句话总结

LINKS 引入了一个包含1亿个单自由度平面连杆机构和110亿条连杆轨迹的数据集，该数据集通过一种新型拓扑算子和高度优化、并行化的仿真生成。该数据集支持数据驱动的逆运动学设计，通过双向切比雪夫距离实现机制检索，可精确匹配目标路径，其检索准确性已在1000万条路径的精选子集上得到验证。

ABSTRACT

In this paper, we introduce LINKS, a dataset of 100 million one degree of freedom planar linkage mechanisms and 1.1 billion coupler curves, which is more than 1000 times larger than any existing database of planar mechanisms and is not limited to specific kinds of mechanisms such as four-bars, six-bars, \etc which are typically what most databases include. LINKS is made up of various components including 100 million mechanisms, the simulation data for each mechanism, normalized paths generated by each mechanism, a curated set of paths, the code used to generate the data and simulate mechanisms, and a live web demo for interactive design of linkage mechanisms. The curated paths are provided as a measure for removing biases in the paths generated by mechanisms that enable a more even design space representation. In this paper, we discuss the details of how we can generate such a large dataset and how we can overcome major issues with such scales. To be able to generate such a large dataset we introduce a new operator to generate 1-DOF mechanism topologies, furthermore, we take many steps to speed up slow simulations of mechanisms by vectorizing our simulations and parallelizing our simulator on a large number of threads, which leads to a simulation 800 times faster than the simple simulation algorithm. This is necessary given on average, 1 out of 500 candidates that are generated are valid~(and all must be simulated to determine their validity), which means billions of simulations must be performed for the generation of this dataset. Then we demonstrate the depth of our dataset through a bi-directional chamfer distance-based shape retrieval study where we show how our dataset can be used directly to find mechanisms that can trace paths very close to desired target paths.

研究动机与目标

• 为平面连杆机构中的数据驱动运动学设计解决大规模、多样化数据集缺乏的问题。
• 克服在生成和模拟大量单自由度平面机构时的计算瓶颈。
• 创建一个全面且无偏见的数据集，覆盖广泛范围的连杆轨迹，以支持逆运动学合成。
• 通过基于度量的相似性搜索，实现高效、可扩展的形状检索，用于路径生成问题。
• 为未来扩展至包含滑块、齿轮等复杂部件奠定基础。

提出的方法

• 引入一种新型拓扑生成算子，系统性地构建最多含20个关节的单自由度平面连杆机构。
• 在数千个线程上实现向量化和并行化仿真，相比朴素仿真实现800倍的加速。
• 生成1亿个机制，每个均模拟完整曲柄旋转一圈以生成连杆轨迹。
• 对所有连杆路径进行归一化，以实现一致的形状比较并减少数据集中的偏差。
• 精选具有代表性的归一化路径子集，以提升设计空间的覆盖度并减少冗余。
• 使用双向切比雪夫距离将查询路径与数据集中的路径进行比较，以实现形状检索。

Figure 1: Two mechanisms where DOF and mobility do not match as a result of redundant linkages. In both images, one of the linkages highlighted in blue has to be removed to remove the redundancies in the mechanism.

实验结果

研究问题

• RQ1 大规模、无偏见的单自由度平面连杆机构数据集能否有效支持数据驱动的逆运动学设计？
• RQ2 简单的搜索策略在多大程度上能成功检索出生成目标路径（数据集中未包含的路径）的机制？
• RQ3 该数据集在多大程度上覆盖了单自由度平面连杆机构中可能的连杆轨迹空间？
• RQ4 该仿真流水线能否在计算效率高的前提下扩展至生成1亿个机制？
• RQ5 精选的路径子集在多大程度上提升了检索性能并改善了设计空间的表征？

主要发现

• 该数据集包含1亿个单自由度平面连杆机构和110亿条连杆轨迹，其规模超过以往任何数据库的1000倍以上。
• 通过向量化和并行化，仿真流水线实现了800倍的加速，使在合理时间内完成数十亿次仿真成为可能。
• 在1000万条路径的子集上，基于简单双向切比雪夫距离的搜索策略成功检索到与三个未见目标路径高度匹配的机制。
• 精选的路径子集有效减少了偏差，并改善了输出空间的表征，从而实现了更可靠的形状检索。
• 该数据集展示了对连杆轨迹类型的广泛覆盖，表明其在逆运动学合成中具有强大潜力。
• 当前版本排除了锁死机构、复杂环路（如双蝴蝶形）以及滑块等非关节部件，这些内容计划在未来扩展。

Figure 2: Overview of the process used to generate planar linkage mechanisms. Note that the initial mechanism used in the topology generator is the simplest 1-DOF mechanism possible. It is important to mention that for each topology, the position generator is run multiple times to obtain multiple ca