System for Flexible & Modular Machine Tending - Economical Automation for Small Batch Sizes
Traditional rigid automation of machine tending is usually not economically viable for small batches and changing products. Within the scope of this project, the concept of a system for flexible machine tending based on a collaborative robot system is being developed and implemented as a prototype. The focus is mainly on the associated flexible safety concept.
The system defines technical contraints by analyzing past medium- to long-term projected product characteristics and batch sizes.Thus the system can adapt in a more modular and flexible manner to workpiece infeed and outfeed, workpiece inspection, workpiece marking, and workpiece post-processing needs. Machine operators should also be able to configure the automation software by drag-and-drop without any programming knowledge. The automation and software framework is also being developed on a modular basis to further integrate machines, controls, and process peripherals in an economically sensible way in the future.
Project partners: Konrad Dummer GmbH | SCHUNK Intec GmbH | FANUC Österreich GmbH | Schmachtl GmbH | Renishaw (Austria) GmbH
DIH2 – A Pan‐European Network of Robotics DIHs for Agile Production
DIH² is an association of 26 Digital Innovation Hubs focused on transforming Small and Medium-sized Enterprises’ (SMEs) production capacities. DIH2 supports SMEs in introducing innovative robotics solutions to respond more flexibly to market requirements and, in particular, to produce small batches more efficiently. Current information can be found at http://dih-squared.eu/.
Currently, the 2nd DIH2 Open Call is accepting submissions for applied projects in industrial robotics and automation. We are happy to support SMEs or Slightly Bigger Companies in consortium formation and project application. DIH2 2nd Open Call
DeepQualityControl – Quality Assurance System Based on Machine Learning
Systems based on conventional image processing are often used for quality assurance of production steps. Their programming is complex, and their use – especially at low cycle times or high production rates – reaches the limits of computing power and reliability.
Deep learning (or neural networks) supports some quality assurance tasks more effectively and tends to be more stable than conventional image processing. Deep learning even offers solutions in cases when it would otherwise be impossible to implement specific tasks. One obstacle is the large number of classified training images required - for example, images with classified defect groups. Since the acquisition of these images requires a great deal of time and effort, the use of Deep Learning is only worthwhile for products with large quantities – an application for small batch sizes is currently not economical.
This project aims to develop a method for the generation of defect classified training images from CAD data of products, including evaluation of the limits and feasibility. Subsequently, suitable deep learning networks will be identified, trained, and their accuracy evaluated. Combining the results in a test installation and applying them to different product families, decision and success factors for quality assurance measures based on Deep Learning are identified and utilized in a decision matrix.
Project partners: ETEC - Automatisierungstechnik Ges.m.b.H. | Maschinenbau Kaindl-Dengg GmbH
Digital Innovation Hub West
The Digital Innovation Hub (DIH) West considers itself a hub and connection point between small and medium-sized enterprises (SMEs) and universities and research partners. DIH West facilitates SMEs' access to digitization know-how and the technological infrastructure of universities and research institutions.The hub aims to build fundamental knowledge, drive innovation, and establish contacts by organizing workshops, events, and working groups. Specializing in the manufacturing, tourism, and software development sectors, DIH West offers services in all phases of digitization - from entry to digital leadership. DIH West is funded by the Federal Ministry for Digital and Economic Affairs, the State of Tyrol, the State of Salzburg, and the State of Vorarlberg. Further information can be found at www.dih-west.at.
Project partners: Universität Innsbruck | Fachhochschule Kufstein Tirol Bildungs GmbH | Fachhochschule Salzburg GmbH | Fachhochschule Vorarlberg GmbH | Fraunhofer Austria Research GmbH | Industriellenvereinigung Tirol | ITG – Innovationsservice für Salzburg | Standortagentur Tirol GmbH | UMIT – Private Universität für Gesundheitswissenschaften, Medizinische Informatik und Technik Gesellschaft mbH | Universität Salzburg | Wirtschafts-Standort Vorarlberg Betriebsansiedlungs GmbH | Wirtschaftskammer Tirol