Graphics Hardware Accelerated Real-Time Machinability Analysis of Free-Form Surfaces

该奖项的研究目标是研究新的几何可制造性/可加工性分析技术，这些技术将利用商用计算机图形硬件的巨大计算能力向设计师产生实时反馈。我们打算在详细设计阶段为设计师提供评估几何可加工性的工具。我们的技术将在离散域中运行，而不是在几何的连续空间实体建模表示上操作，从而简化计算并减少数值误差。这些方法将利用图形处理单元(GPU)的能力来处理使用离散域方法时出现的大型数据集。结果将是作为专用过滤器实现的各种可加工性查询的集合，这些过滤器将作用于零件的镶嵌模型(通常用于三维显示)。可制造性评估将包括在修改的图形流水线内使用适当的过滤器查看自由曲面。如果成功，这项研究将产生类似于标准图形流水线的可制造性评估流水线，该流水线将被集成到现代CAD系统中用于实时几何可加工性评估。设计人员将能够在标准图形显示和可加工性评估显示之间切换，以便在详细设计期间直观地识别设计中的问题区域。作为教育计划和推广的一部分，我们将开发可视化工具，这些工具将用于课程，以提高学生对几何加工性的理解。这些插件将用于计算机辅助设计的本科生课程和K-12推广计划。来自计算机科学(程序员)和机械工程(测试部件开发人员)的有前途的本科生将通过研究学者计划(RSP)招募，使用REU资金。更广泛的社会影响是通过这项研究对保持美国产品设计师在全球市场上的竞争力的影响来解决的。

The research objective of this award is to investigate new geometric manufacturability/machinability analysis techniques that will leverage the massive computational capabilities of commodity computer graphics hardware to generate real-time feedback to designers. We intend to provide designers with tools to evaluate geometric machinability during the detailed design phase. Instead of operating on the continuous-space solid modeling representations of geometry, our techniques will function in the discretized domain thus simplifying calculations and reducing numerical errors. These methods will exploit the capabilities of graphical processing units (GPUs) to handle the large data-sets which occur when using discrete domain methods. The result will be a collection of various machinability queries implemented as specialized filters that will act on tessellated models of the parts (commonly used for three dimensional display). Manufacturability evaluation will consist of viewing the free-form surfaces with the appropriate filters within the modified graphics pipeline.If successful, this research will result in manufacturability evaluation pipelines analogous to the standard graphics pipeline which will be integrated into contemporary CAD systems for real-time geometric machinability evaluations. Designers will be able to switch between the standard graphics display and the machinability evaluation display to visually identify problem areas in the design during detailed design. As part of the educational plan and outreach, we will develop visualization tools that will be used in courses to improve students' understanding of geometric machinability. These plug-ins will be used in an undergraduate course in computer aided design and in K-12 outreach programs. Promising undergraduate students from computer science (as programmers) and mechanical engineering (test part developers) will be recruited through the Research Scholars Program (RSP) using REU funding. Broader societal impact is addressed through the effect this research will have on maintaining the competitiveness of US product designers in the global market.