Résumé: This paper reports the design and implementation of a system for intelligent extraction of manufacturable features from ?engineering drawings? created in CADD format using the techniques of string based syntactic pattern recognition and natural language processing. Drawing entities are processed to derive feature string patterns, which are syntactically analyzed to detect the feature topology and geometry. The information of the features so detected is augmented by non-geometric data like dimensions, position, tolerances and the relevant drawing callouts, and represented as a Part Model using object oriented design paradigm (OOD). The system is capable of detecting generic classes of features like holes, pockets, steps, slots, bosses, etc., on prismatic machined components. The informationally complete part model can be easily linked to the downline CAD/CAM activities in the product cycle. The system, thus, demonstrates the feasibility of design automation from CADD models.

Langue: Anglais