逆向工程中曲面自动重建的柔性解决方案研究

Reverse engineering is an important technique for innovative product design and rapid response manufacturing. Coordinate measuring machines have made great progress in recent years. Cloud points, which contain more detail of the scanned objects, can be obtained conveniently. But cloud data bring challenge to data processing and model reconstruction technique. Software for reverse engineering fell behind corresponding hardware for a time. Digital model reconstruction became a bottleneck problem in reverse engineering. In this project, we studied the following aspects in reverse engineering: (1) digitized points preprocessing, especially cloud data compression;(2) triangular mesh surface reconstruction from scattered points;(3) error analysis for the reconstructed mesh surface;(4) error controllable mesh simplification; (5) hole repairing for the mesh surface;(6) generalized surface spline; (7) quadrilateral patches network segmentation and piecewise B-spline surfaces fitting on arbitrary triangle mesh. A reverse engineering software prototype was developed on the basis of algorithmic research. The software has been used in some enterprises to reconstruct typical industrial products, including parts on planes, car moulds, motorcycles, parts on diesel engines etc. The methods accomplished in this project can be widely used in digital model reconstruction of existing objects, innovative design of products with high aesthetics requirements, error inspection for products with complex surfaces, large scale customized manufacturing, GIS, Web-based 3D product distribution etc.

根据坐标测量机迅速发展给曲面重建提出的新课题，本项目以任意散乱点集为处理对象，首先自动重建曲面的三角化逼近模型，再对三角化网格进行子波变换，以实现精度完全可控的任意层次的模型简化。通过对三角网格进行自动四边界区域划分，建立参数化映射，产生整体光滑的样条曲面拟合。本项目对改变逆向工程软件明显落后于硬件的现状具有重要意义。