Geometric Algorithms for Recognition of Features from Solid Models

Collaborative engineering has expanded the scope of traditional engineering design to include the identification and elimination of problems in the manufacturing process. Manufacturing features and feature-based representations have become an integral part of research on manufacturing systems, due to their ability to model the correspondence between design information and manufacturing activities. One necessary component of an integrated Computer Aided Design/Computer Aided Manufacturing (CAD/CAM) environment is a tool to automatically recognize manufacturing features from a CAD or solid model. In this thesis we present a methodology for recognizing a class of machining features and for addressing the computational issues involved in building tractable and scalable solutions for automated feature recognition. This approach is described for a class of volumetric features based on material removal volumes produced by operations on 3-axis vertical machining centers. A computational framework is developed for representing different types of common machining features and specifying the recognition problem. Based on this framework, novel serial and multi-processor recognition algorithms are described and analyzed with respect to their completeness and complexity. The goal of this dissertation is to advance the understanding of the basic computational issues that arise in feature recognition from solid models of mechanical artifacts and to facilitate development of effective and efficient systems that can scale to address industrial problems.