SGER: Flexible Active Computer Integrated Inspection Based on Computer-Aided Design Models

[9319208 . Marefat]该项目的目标是开发一个新的研究项目，涉及主动机器视觉、柔性制造和人工智能方面的多学科工作，解决基于（CAD）设计模型的制造（特别是加工）部件的自动视觉检查中的CAD问题。本研究涉及两个方面，这两个方面对于在这个新的研究方向上取得成功都很重要。第一个方面涉及自动智能生成检查计划。这一方面可以离线完成，本质上构建了检查策略。第二个方面涉及用于有效执行所生成的检查计划以及传感器和其他资源的较低级方法。为了实现综合智能检测的目标，本项目解决了几个相互关联的问题。这些问题包括：(i)开发分层表示机制，以有效捕获有关几何实体及其关系、传感器和计划的知识；（ii）推理机制，以确定待检查对象的不同特征的不同属性，以及可用于检查每个属性的替代策略；（iii）摄像机的放置和视角策略，以及确定每种配置中的可见实体的策略。（四）构建方案的优化。这一问题的有效解决方案显然是非常重要的，不仅从自动化的角度来看，而且为了有效地收集和使用传感器信息，以克服所需数据处理的数量和复杂性，以获得足够的性能。所要求的支助将用于调查，以取得进一步的结果，并开发一个有限的原型系统。最终目标是向工业客户展示开发的有限原型，以激发他们的兴趣和参与，并将研究结果记录在专业出版物中。

9319208 Marefat The objectives of this project are to develop a new research program involving multi disciplinary work in active machine vision, flexible manufacturing, and artificial intelligence, which address CAD problems in automated visual inspection of manufactured (particularly machined) components based on their (CAD) design models. This research addresses two facets which are both important for success in this new research direction. The first facet is concerned with automated intelligent generation of inspection plans. This facet, which may be done off line, essentially constructs the inspection strategy. The second facet deals with the lower level methods for effective execution of the generated inspection plan together with sensors and other resources. In order to achieve the integrated intelligent inspection goals, the project addresses several interrelated problems. These problems include: (i) developing hierarchical representation mechanisms to effectively capture the knowledge about geometric entities, their relationships, sensors, and plans (ii) reasoning mechanisms to determine the different attributes of the different features of object to be inspected, and the alternative strategies which can be used for inspection of each attribute, (iii) strategies for placement and viewing angles of cameras, and for determining the visible entities in each configuration, and (iv) optimization of the constructed plan. An effective solution to this problem is clearly very important not only from the automation point of view, but also in order to effectively gather and use sensor information to overcome the volume and complexity of the needed data pro cessing for adequate performance. The requested support will be used for investigations to achieve further results, and to develop a limited prototype system. The ultimate goals are to demonstrate the developed limited prototype to industrial clients to stimulate their interest and involvement, and also to document the research results in professional publications.