CAREER: Geometric Reasoning for Large Engineering Knowledge-Bases

职业：大型工程知识库的几何推理

• 批准号: 9733545
• 负责人: William Regli
• 金额: $ 35.5万
• 依托单位: Drexel University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-10-01 至 2003-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9733545&HistoricalAwards=false
• 关键词: CAREER; Geometric; Reasoning; Large; Engineering

This research develops representations and algorithms to manage large digital libraries of engineering knowledge. In engineering design, it is conservatively estimated that more than 75% of design activity is case-based. This research identifies new computational methods for reasoning about complex geometric and engineering information. It develops the mathematical foundation and algorithmic tools to support content-based retrieval from large engineering knowledge-bases (such as are generated by modern CAD systems). This research identifies: (1) classes of heuristic search algorithms for automated indexing and retrieval of models from design knowledge-bases; (2) data structures for defining arbitrary relationships among engineering data elements, based on their relationships through product model geometry; (3) search interfaces for supporting customized queries from outside agents. Additionally, this project develops an educational plan that integrates the core computer science curriculum with complex problems drawn from real-world application domains. This is accomplished through incorporating interdisciplinary problem solving into basic Computer Science classes, outreach activities with industry, and the National Design Repository---digital textbook of design knowledge for users around the globe. This CAREER plan will have an extensive impact on engineering design and computer-aided manufacturing. http://www.mcs.drexel.edu/ wregli/EKB

这项研究开发了表示和算法来管理大型工程知识数字图书馆。 在工程设计中，保守估计75%以上的设计活动都是基于案例的。 这项研究确定了用于推理复杂几何和工程信息的新计算方法。 它开发了数学基础和算法工具，以支持从大型工程知识库（例如由现代 CAD 系统生成的知识库）中进行基于内容的检索。 这项研究确定了：（1）用于从设计知识库中自动索引和检索模型的启发式搜索算法类别； (2) 数据结构，用于根据工程数据元素通过产品模型几何形状的关系来定义工程数据元素之间的任意关系； (3) 搜索接口，支持外部代理的定制查询。 此外，该项目还制定了一项教育计划，将核心计算机科学课程与来自现实世界应用领域的复杂问题相结合。这是通过将跨学科问题解决纳入基础计算机科学课程、行业推广活动以及国家设计知识库——为全球用户提供的设计知识数字教科书来实现的。 该职业计划将对工程设计和计算机辅助制造产生广泛影响。 http://www.mcs.drexel.edu/wregli/EKB