Process modeling of non-contact reverse engineering process

In presented paper a design model is developed which is used to find an optimal design space in order to improve precision and accuracy of reversed 3D models. Significant parameters and their mutual effects in scanning process are recognized with the help of Design of Experiments (DOE) principles, and in second step recognized significant parameters are also optimized using the RSM (Response Surface Methodology). Back propagated neural network is used for prediction of response factors whose input variables are not possible to set during experimentation due to the nature of the scan device. With neural network (NN) predicted response-design data, the Analysis of Variance (ANOVA) test was performed. On a basis of significant parameters the analytical model is established, which enables prediction and optimization of critical scan parameters, according to the prescribed standard deviation (abberation) of analysed 3D CAD models. After getting knowledge about the relationships between significant parameters it is possible to predict and optimize them using a model established in the RSM on a basis of D-optimal Design of Experiment test (DOE). At the end also the precision assessment of predicted response factors by DOE approach and by neural network approach is done and results are beeing compared.

[1]  Andreas Koschan,et al.  3D CAD model generation of mechanical parts using coded‐pattern projection and laser triangulation systems , 2005 .

[2]  Lijiang Wang,et al.  A study of the precision factors of large-scale object surface profile laser scanning measurement , 2002 .

[3]  Andrew W. Fitzgibbon,et al.  Calibration, data consistency and model acquisition with laser stripers , 1998, Int. J. Comput. Integr. Manuf..

[4]  Ken R. McNaught,et al.  A comparison of experimental designs in the development of a neural network simulation metamodel , 2004, Simul. Model. Pract. Theory.

[5]  Elizabeth A. Croft,et al.  Machine vision system for curved surface inspection , 2000, Machine Vision and Applications.

[6]  Heinrich Müller,et al.  Interpolation and Approximation of Surfaces from Three-Dimensional Scattered Data Points , 1997, Scientific Visualization Conference (dagstuhl '97).

[7]  Shiuh-Jer Huang,et al.  A three-dimensional non-contact measurement system , 1997 .

[8]  Jose C. Principe,et al.  Neural and Adaptive Systems: Fundamentals through Simulations with CD-ROM , 1999 .

[9]  Jože Duhovnik,et al.  Case Study – surface reconstruction from point clouds for prosthesis production , 2007 .

[10]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[11]  Gou-Jen Wang,et al.  Reverse Engineering of Sculptured Surfaces by Four-Axis Non-Contacting Scanning , 1999 .

[12]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[13]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[14]  V. Kvasnicka,et al.  Neural and Adaptive Systems: Fundamentals Through Simulations , 2001, IEEE Trans. Neural Networks.

[15]  F J Shiou,et al.  Development of a non-contact multi-axis reverse engineering measurement system for small complex objects , 2005 .