Re-parameterization for Solid Modeling
Date of Completion
My dissertation formulates novel theoretical and computational tools to create and manipulate 3-dimensional solid models of mechanical parts. Specifically, the first set of tools allows useful geometric editing operations of solid models in a manner that is independent of any existing parameterization, which circumvents some of the main difficulties encountered by the commercial CAD packages relying on parametric modeling technology. Further, the second set of tools will focus on the important problem of automatically reconstructing digital geometry of mechanical parts from scanned data, which will include parametric definitions of design and manufacturing features. The proposed novel approach will, in turn, circumvent the heavy and extremely time consuming user interaction required by the currently available methods, and will have a dramatic impact on the robustness and accuracy of the resulting feature based solid models. My recent work has shown that the proposed approach to convert point clouds directly into parameterized feature based solid models is practicable and effective. Some problems that I will explore include: · Robustly building topological maps of faces adjacency; the faces are extracted from the original point cloud; · Effectively searching and extracting largest point cloud subsets corresponding to individual features by feature recognition; · Efficiently fitting parametric solid feature templates into the point cloud subsets; · Accurately controlling position and orientation of features for Boolean operations. ^
Wu, Nier, "Re-parameterization for Solid Modeling" (2011). Doctoral Dissertations. AAI3504584.