Accuracy analysis of 3D data collection and free-form modelling methods

Abstract

In industrial applications such as the automotive industry, it is often necessary to design and manufacture surfaces to a desired smoothness. This can be achieved by computer-aided design (CAD) methods, or from the creation of a physical prototype. To gain an adequate feel for the part to be manufactured, physical prototype models are often made by the designer.In the case of free-form surfaces, it can be difficult to judge the overall smoothness of the surface. The optimum smoothness, or continuity, relies on there being no continuity deviations. Without a mathematical model of the surface, it is difficult to confirm that a manufactured surface conforms to the desired specifications. The only method is by way of a designer’s judgement and experience.Thus, to accurately assess a physical surface it is necessary to inspect the part. This is with the aim of collecting accurate 3D data, allowing a 3D CAD surface to be generated. This forms a quality assessment stage. However, this method of verification can also succumb to inaccuracies.This paper offers a method of automating the verification of an acceptable free-form surface, using the co-ordinate measuring machine (CMM) as a means of collecting data from the surface in question. Methods are proposed to analyse the surface for optimum continuity, using computer-aided geometric design (CAGD), as well as assess the CMM data accuracy. This surface can be improved, if desired, using reverse engineering (RE) methods.