Coping with the complexity and infinite variety of geometric features : a cooperative geometric feature recognition system
處理幾何特徵之無限及複雜性 : 具合作能力之幾何認徵系統
Student thesis: Doctoral Thesis
Integrating various manufacturing activities by means of concurrent paradigm have proved to be one of the promising concepts to achieve successful manufacturing environment. Concurrent engineering demands a high degree of integration for various computer-aided systems directed at manufacturing activities. These activities, from design, planning to actual production process, need to work cooperatively in a seamless Computer Integrated Manufacturing (CIM) environment. Due to this reason, much research effort has been directed towards integrating Computer-Aided Design and Computer-Aided Manufacturing (CAD/CAM) with the Computer-Aided Process Planning (CAPP) systems in the past decade. However, planning of manufacturing process from a designed product requires a high level of human reasoning, especially during the interpretation of geometric features. It becomes an important task to have a generic geometric feature recognition system for this integration process. In this research, a fully algorithmic method for geometric feature recognition (GFR) is developed. A feature taxonomy based on the concept of atomic features (called Primitive Template Features, PTFs) and their variations (Variations of Primitive Template Features, VPTFs) enables the recognition of the infinite variety of features that arise in practice from a finite feature database. Computational economy is achieved by separating the feature extraction and feature identification phases and utilizing new algorithms for decomposing protrusion and depression types of complex features in terms of constituent PTF/VPTFs. The algorithms are based on a heuristic understanding of the ways in which PTFs interact. Features are represented by a newly developed coding system that embed the topological and coarse geometry information of the feature. Furthermore, a Feature Coding Algebra (FCA) is developed to analyze the constitution and the machining sequence of the complex interacting feature. The determination of intermediate features can facilitate the manufacturing planning process. In order to integrate the GFR system with other manufacturing systems, a holonic approach is used while developing the system design methodology. The modularity and cooperative characteristics facilitate the maintenance and expandability of the GFR system. In order to demonstrate the agility of the algorithms, a software based on an object-oriented approach has been developed for the implementation of the cooperative GFR system. Issues concerning how one may design a cooperative GFR system are also discussed.
- Pattern recognition systems