Perfectly coated surfaces are an essential quality feature in the automotive and consumer goods industries. They are the result of an optimized, controlled coating process. Because entire assemblies could be rejected if Out-of-Specification (OOS) parts are installed, this has a severe economic impact. This paper presents a novel, line-integrated multi-camera system with intelligent algorithms for anomaly detection on small KTL-coated aluminum parts. The system also aims to automatize the previously used human inspection to a sophisticated and automated vision system that efficiently detects defects and anomalies on coated parts.

Zusammenfassung: Eine Möglichkeit zur Vorhersage und damit der Korrektur thermo-elastischer Verformungen liegt in der Multivariaten Regressionsanalyse (MRA). Kennfelder, als Vertreter der MRA, ermöglichen die direkte, nicht-lineare Abbildung von einem kleinen Satz von Temperatursensoren auf der Werkzeugmaschine auf ihre zugehörige Verlagerung. Das Kapitel beschreibt die Berechnung derartiger Korrekturkennfelder aus einer Trainingsdatenbank und zeigt deren Anwendung und Güte am Beispiel eines Fünf-Achs-Bearbeitungszentrums.

Flexibility and individuality in product portfolios are increasingly important in manufacturing, necessitating highly adaptable production systems. Industrial robots, despite their potential, are often underutilized due to limited autonomy and programming challenges. This paper presents a modular software toolbox designed for cognitive robotics while addressing the research question of how a modular software toolbox can enhance the cognitive capabilities and autonomy of industrial robots to improve adaptability and efficiency in automated manufacturing processes, particularly in the context of small to medium batch inspections, alongside the increase in efficiency during commissioning and application. The toolbox features reconfigurable algorithms for object recognition, handling, and quality inspections using machine learning methods such as CNNs and SVMs. Optimized for standard industrial computers, the system ensures efficient resource use, achieving object recognition at 20 frames per second. The setup includes a six-axis robot with a CNC controller and high-precision sensors, offering significant cost and time savings for small- to medium-batch inspections. Validation demonstrates the toolbox's potential to enhance flexibility and intelligence in manufacturing systems.