Point cloud based design and manufacturing

点击翻译按钮获取中文摘要

Conventional CAD/CAM software uses simple "CSG" block, etc. primitives combined with Boolean intersect, etc. operations to create 3-D models.  Sculptured B-spline curves and surfaces can be added, and "evaluated" to yield a topological set of "B-rep" bounded, volume enclosing surfaces.  In practice, the CSG to B-rep evaluation process involves inexact surface/surface intersections, and floating point round-off based exchange reliability problems remain.  Concurrently, stereo vision and laser triangulation systems, used to digitize design concept models, etc. into dense point clouds, have grown in acceptance.  These point clouds are converted to fitted B-spline surfaces and "faired" to smooth out curvature "bumps".  Use of B-splines reduces data storage requirements, and is compatible with the existing IGES/STEP exchange standards.     Ultimately, however, CAD models are converted back to point style representations.  Graphics hardware is based on rapid display of polygons that are created by "tesselating" from the B-rep model.  For manufacturing, Rapid Prototyping machines require STL triangle based file representations.  Ultimately, Computer Numerical Control (CNC) machining requires creation of tool paths that consist of (short) linear motions.  Their inherent reliability, combined with ever increasing computer speed and memory availability, has created a CAD research community trend away from CSG/B-rep and B-spline models to wholly point cloud representations.  The proposed research will harness this emerging trend, and extend early but successful applicant research from stereo vision and laser triagulation based sheet metal strain analysis to more general surface gray level and color measurement, point cloud based collision detection, and rapid prototype surface coloring using a 3-D Coordinate Measuring Machine (CMM) mounted inkjet printer head.  Additional potential applications include electrochemical etching for sheet metal and copper printed circuit boards, Computer Aided Tissue Engineering (CATE) by inkjet cell deposition, and haptic surgery planned using medical Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scan data.