Fraunhofer-Publica

The Fraunhofer-Publica has been successfully documenting the research results of the Fraunhofer-Gesellschaft for over 30 years. The platform enables the collaborative linking of research-relevant objects and disseminates within the international scientific community.

The Fraunhofer-Publica thus fulfils its responsibility to promote the transfer of knowledge and know-how to industry and society.

Categories

266479

Research outputs

As an application-oriented research organisation, Fraunhofer aims to conduct highly innovative and solution-oriented research - for the benefit of society and to strengthen the German and European economy.

13927

Projects

Fraunhofer is tackling the current challenges facing industry head on. By pooling their expertise and involving industrial partners at an early stage, the Fraunhofer Institutes involved in the projects aim to turn original scientific ideas into marketable products as quickly as possible.

7739

Researchers

Scientific achievement and practical relevance are not opposites - at Fraunhofer they are mutually dependent. Thanks to the close organisational links between Fraunhofer Institutes and universities, science at Fraunhofer is conducted at an internationally first-class level.

77

Institutes

The Fraunhofer-Gesellschaft is the leading organisation for applied research in Europe. Institutes and research facilities work under its umbrella at various locations throughout Germany.

Recent Additions

  • Publication
    Effizientere Lackierprozesse durch selbstlernende Verhaltensmodelle
    ( 2023)
    Tiedje, Oliver
    ;
    Hecker, Meiko
    Die Zukunft des Lackierprozesses ist Thema eines aktuellen gemeinschaftlichen Forschungsprojekts. Das Ziel: Fehlerquoten und Stillstandzeiten reduzieren, den Lackverbrauch verringern und die Anlaufzeit für neue Farben verkürzen. Die beteiligten Projektpartner konnten die durchgängige Vernetzung der Prozess- und Qualitätsdaten in einem Demonstrator erfolgreich umsetzen.
  • Publication
    Robot Control for Assembly Task by Imitation Learning
    ( 2023-02-03)
    Kandur, Akhil Teja
    ;
    ten Hompel, Michael
    ;
    Gouda, Anas
    ;
    Nguyen, Trinh Cong
    Robots and automated systems are used in the industrial production sector to sort, put together, and construct a variety of complicated mechanical structures. The robots are capable of being programmed to carry out several tasks during any part assembly process. However, because manipulators need a similar amount of human labor to program a given set of robotic actions, the complicated programming effort is not scaled in relation to small batch manufacturing as opposed to big volume production. It is necessary to train the manipulator for extremely repetitious activities, but it is not cost-effective to program a robot for each individual unit of production, such as the assembly of a unit product or the creation of multiple bespoke products. While there are several methods to help a robot complete a task more quickly and simply, imitation learning will be used in this instance. More specifically, it will be learned how to control the action of a collaborative robot arm with a gripper to carry out the assembly of basic parts like wooden blocks. The hardware consists of a Franka Emika Panda collaborative robot arm with a controller, mounted gripper, and a camera for monitoring the workspace. The camera works in conjunction with existing object-tracking software to identify the scene and track the 6D pose of the scene’s necessary elements. While the detection and tracking module captures the relative 6d poses of the components, the assembling procedure must be shown by a human or robotic (human-guided) assistant. To learn a general movement that will converge to the desired goal location, motion primitives are used. It is possible to generalize this learning so that the robot learns to reach various goal positions while adhering to the limitations of the displayed trajectory. Additionally, multiple features can be implemented with different parameters set to the learning. The object recognition and tracking module will once more be a component of the control loop for the execution, i.e., replication of the displayed assembly process, and motion primitives will be utilized to produce generalized trajectories tailored to the present scene.
  • Publication
    Automatisierungstechnologien in der Intralogistik
    ( 2023-11-06)
    Minkhofer, Stephan
    ;
    ten Hompel, Michael
    ;
    Barlang, Maximilian
    In this paper, we will first look at the fundamental drivers that are pushing logistics to use more automation technologies. Technologies in the various functional areas of a warehouse are then presented. Technical data and the functionalities of the technologies are described. On this basis, a strengths and weaknesses analysis of the technologies is carried out. In order to be able to assess the maturity level of the technologies, the technologies are assigned to the maturity levels based on criteria from the method. This is followed by a process-related analysis of incoming goods as an example process. The aim here is to identify potential errors that may occur. In the final step, the possibility of retrofitting is examined using the information obtained from the previous analyses.
  • Publication
    Entwicklung eines Tools zur Vorauswahl von Kommissioniersystemen mit Hilfe von Visual Basic for Application
    ( 2023-04-28)
    Birgith Niesert, Simone Caroline Lucia
    ;
    ten Hompel, Michael
    ;
    Niemann, Friederich
    Das Ziel diese Masterarbeit ist es, ein Tool zur Vorauswahl von Kommissioniersystemen mit Hilfe von Visual Basic for Application zu entwickeln. Dazu sollen unterschiedliche Kommissioniersysteme in einer Datei anhand der Kosten, Leistung, Lagergröße, Anzahl der Lagereinheiten und Anzahl der einzelnen Kommissioniersysteme vergleichbar gemacht werden. Um Kommissioniersysteme zu vergleichen, werden die Vor- und Nachteile einiger Systeme vor der Entwicklung des Tools anhand von Literaturrecherchen zusammengetragen. Aufgrund der Vielzahl an Kombinationsmöglichkeiten von Kommissioniersystemen werden in das zu entwickelnde Tool keine Systeme voreingetragen. Diese können durch den Anwender selbst integriert werden. Dafür besteht einerseits die Möglichkeit, das System über Grundbewegungen einer MTM-Analyse darzustellen, wobei die Daten aus der MTM-Analyse durch das Unternehmen selbst eingetragen werden müssen. Weiter kann ein System anhand eines Lagers beschrieben werden. Die Kommissionierzeiten der Systeme werden anhand von Einflussgrößen berechnet, aus denen dann wiederum die Vergleichsgrößen berechnet werden. Anhand der Vergleichsgrößen kann der Anwender eine Vorauswahl der Kommissioniersysteme treffen, die dann in der Detail- und Feinplanung detaillierter geplant werden können. Bei der Vorauswahl ist zu beachten, dass keine direkte Entscheidung aus dem Tool ablesbar ist. Es werden nur ein Teil der Einflussgrößen betrachtet, da einige Einflussgrößen nicht messbar sind und daher von dem Anwender bewertet werden müssen. Weiterführend kann das Tool hinsichtlich der Einflussgrößen ausgebaut werden. Außerdem besteht die Möglichkeit in einer weiteren Arbeit die Vergleichsgrößen durch eine Simulation validieren zu lassen. Dies trägt dazu bei, dass die Berechnungen genauer werden und das Tool in der Praxis noch besser und verlässlicher eingesetzt werden kann.

Most viewed

  • Publication
    Hierarchical Clock Synchronization in MPI
    ( 2018)
    Hunold, S.
    ;
    Carpen-Amarie, A.
    MPI benchmarks are used for analyzing or tuning the performance of MPI libraries. Generally, every MPI library should be adjusted to the given parallel machine, especially on supercomputers. System operators can define which algorithm should be selected for a specific MPI operation, and this decision which algorithm to select is usually made after analyzing bench-mark results. The problem is that the latency of communication operations in MPI is very sensitive to the chosen data acquisition and data processing method. For that reason, depending on how the performance is measured, system operators may end up with a completely different MPI library setup. In the present work, we focus on the problem of precisely measuring the latency of collective operations, in particular, for small payloads, where external experimental factors play a significant role. We present a novel clock synchronization algorithm, which exploits the hierarchical architecture of compute clusters, and we show that it outperforms previous approaches, both in run-time and in precision. We also propose a different scheme to obtain precise MPI run-time measurements (called Round-Time), which is based on given, fixed time slices, as opposed to the traditional way of measuring for a predefined number of repetitions. We also highlight that the use of MPI_Barrier has a significant effect on experimentally determined latency values of MPI collectives. We argue that MPI_Barrier should be avoided if the average run-time of the barrier function is in the same order of magnitude as the run-time of the MPI function to be measured.
  • Publication
    Characterization of self-cleaning properties on superhydrophobic aluminum surfaces fabricated by direct laser writing and direct laser interference patterning
    ( 2020)
    Milles, Stephan
    ;
    Soldera, Marcos
    ;
    Kuntze, Thomas
    ;
    Lasagni, Andrés-Fabián
    Self-cleaning ability on technical surfaces can increase the added value of a product. A common path to achieve this property is making the surface superhydrophobic so that water droplets can roll down, picking up dirt particles. In this contribution, the self-cleaning efficiency of Al surfaces structured with direct laser writing (DLW), direct laser interference patterning (DLIP) and a combination of both technologies was quantitatively determined. This was performed by developing a characterization method, where the treated samples are firstly covered with either MnO2 or polyamide micro-particles, then tilted by 15° and 30° and finally washed applying up to nine water droplets (10 µl) over the contaminated surfaces. Then, an optical analysis by image processing of the remaining contamination particles on the textured surfaces was realized after each droplet rolled over the surface. The DLIP textures showed the best performance, allowing the removal of more than 90% of the particles after just three droplets were released. High-speed videos and scanning electron microscopy characterization allowed a deeper understanding on the cleaning behavior and on the relationship between surface microstructure and particle size and shape.