Peter Rummel

Model Generation for Sensor-Guided Flexible Assembly Systems

Abstract Current research in sensor technology is leading to task- independent algorithms and knowledge- based processing systems. The paper emphasizes that in complex production processes, not only the geometric object descriptions but also the peripheral conditions such as cell- layout, surface properties, sensor interdependencies, and data integration must be considered. An adequate definition of the term ‘sensor model’ is given. By using AI- techniques, a generic knowledge- based machine vision system is designed that consists of four main elements: a declarative part, a generative element, a rule- based component, and an application dependent analysis program. The principles above are discussed in terms of two implemented 2D/3D- sensors. Finally an evaluation of prototype installations is given.

A Model-Based Grey-Scale Sensor System for Robotics

Summary Production devices for fabrication of medium and small aeries have to be flexible to meet the changing conditions of production. Scene- analysis systems for the recognition and inspection of workpieces are an important tool for flexible automation. Most vision systems currently available for robotics show severe limitations concerning recognition time, workpiece acquisition, illumination etc. This paper describes a Grey- Scale Sensor System for recognition of touching and overlapping workpieces, designed to meet the requirements of real applications in the factory. The system comprises 2 stages: a feature- extraction stage, which is designed mainly in special- purpose hardware and a model- based analysis stage realized in software. The system works with edge- based geometrical primitives and surface features. The model comprises topographical and procedural information in order to control the image- analysis, which is based on a Heuristic Search technique. A prototype version has been finished and interfaced to a robot. First experiences show that the system is working satisfactoryly in a wide range of environmental conditions and with different kinds of workpieces.

GSS—A FAST, MODEL-BASED VISION SYSTEM FOR WORKPIECE RECOGNITION

This paper describes a new visual sensor system for the recognition of touching and overlapping workpieces, designed to meet the requirements of real applications in the factory. The system analyses the grey scale image and comprises two stages: a feature-extraction stage, which is designed mainly in special-purpose hardware, and a model-based analysis stage. The system works with edge-based geometrical primitives and surface features. The model comprises topographical and procedural information in order to control the image-analysis, which is based on the “heuristic-search” technique. The hardware and software design is modular. The system is working well in a wide range of environmental conditions and with different kinds of workpieces. A prototype version is completed and has proved its reliability and performance in several applications in the factory.