CHS: Small: Printable Partitioning of 3D Models using Level Set Methods

Ohio State University, The, Columbus OH

Investigators

Abstract

As 3D printing technology becomes mainstream and starts to revolutionize our daily lives as well as the manufacturing industries, a critical problem is about to emerge: how can we effectively divide an object into multiple printable components, when the design of this object does not provide such a partitioning? We might need such a partitioning because the object is too large to fit into the printer, or we want to reduce the printing time, or the object needs to be packed more tightly, or we just want to make part of the object replaceable or reusable. In this project, the PI will investigate object partitioning from six aspects: printability, stress load, surface details, packing, assembly, and consistency among multiple objects. The computational tools developed from this research will broaden the use of 3D printing technology in a wide range of areas. the PI's existing collaborations with industrial partners such as Adobe and NVIDIA will enable project outcomes to have broad impact, while the PI will publish demo applications and make source code available online to further accelerate dissemination. The PI will also collaborate with his institution's University Center for the Advancement of Teaching (UCAT) on incorporating the results of this research into both undergraduate and graduate graphics courses. The ultimate goal is to develop an automatic partitioning system that quickly converts the mesh of one object into a series of meshes, each of which represents an object component. Preliminary results indicate that the level set framework is a promising approach. The project encompasses a comprehensive research agenda focused on three fundamental subproblems: how to analyze the quality of a partitioning according to a variety of printing, packing, and assembly criteria; based on the previous analysis, how to improve the quality of a partitioning using level set methods; and how to facilitate the practical use of a partitioning by constructing auxiliary structures. Many of the subproblems central to this research are new and have not been systematically studied before. The development of this technology will require both theoretical and practical studies of system integration, computational efficiency, and time-stepping. Finding solutions to such issues within the level set framework will allow the PI to formulate his system as a localized iterative partitioning optimization process, whose result can meet the requirement of different 3D printing applications.