MACHINE INTERPRETATION OF CAD DATA FOR MANUFACTURING APPLICATIONS

Machine interpretation of the shape of a component from CAD databases is an important problem in CAD/CAM, computer vision, and intelligent m anufacturing. It can be used in CAD/CAM for evaluation of designs, in computer vision for machine recognition and machine inspection of obje cts, and in intelligent manufacturing for automating and integrating t he link between design and manufacturing. This topic has been an activ e area of research since the late '70s, and a significant number of co mputational methods have been preposed to identify portions of the geo metry of a part having engineering significance (here called ''feature s'').(1) However, each proposed mechanism has been able to solve the p roblem only for components within a restricted geometric domain (such as polyhedral components), or only for components whose features inter act with each other in a restricted manner. The purposes of this artic le are to review and summarize the development of research on machine recognition of features from CAD data, to discuss the advantages and p otential problems of each approach, and to point out some of the promi sing directions future investigations may take. Since most work in thi s field has focused on machining features, the article primarily cover s those features associated with the manufacturing domain. In order to better understand the state of the art, methods of automated feature recognition are divided into the following categories of methods based on their approach: graph-based, syntactic pattern recognition, rule-b ased, and volumetric. Within each category Re have studied issues such as the definition of features, mechanisms developed for recognition o f features, the application scope, and the assumptions made. In additi on, the problem is addressed from the perspective of information input requirements and the advantages and disadvantages of boundary represe ntation, constructive solid geometry (CSG), and 2D drawings with respe ct to machine recognition of features are examined. Emphasis is placed on the mechanisms for attacking problems associated with interacting features.