EXTRACTION OF FEATURE INFORMATION FROM THREE-DIMENSIONAL CAD DATA
The future success of CAD and CAM depends on the ability of these two processes to communicate with each other and with the intermediate manufacturing database. CAD-generated objects can be defined and stored as complete geometric and topologic solids. The elements of CAM, however, do not make full use of the part description because it exists in terms of low-level faces, edges and vertices or primitive volumes unrelated to the manufacturing planning task. Consequently, manufacturing planning still depends upon human expertise and input to interpret the part definition according to manufacturing needs. The automatic linking of CAD and CAM involves a re-interpretation of the part database to extract manufacturing-specific semantic knowledge about the part. This knowledge can enable process planning and other CAM operations to proceed automatically without human guidance or intervention. A method has been developed to extract semantic knowledge useful to manufacturing in the form of part features. The procedure consists of searching the part description, recognizing cavity features, extracting those features as solid volumes of material to be removed and arranging them in a feature graph, a high level data structure more appropriate to manufacturing process planning. The feature recognition step uses logic programming implemented in PROLOG and expert system techniques. Using rules of logic, definitions have been formulated to describe some common features of a part, such as holes and slots. This work has relevance in the areas of CAD/CAM linking, automated process planning, expert systems and, in fact, the interpretation of any 3-D data.
Off-Campus Purdue Users:
To access this dissertation, please log in to our