ERF Workshop 2020: AppliedAI in Agile Production


Dr. Michael SUPPA, Roboception GmbH, Germany

Prof Markus Vincze, Technical University of Vienna, Austria


Motivation and Objectives:

Perception is one of the key technologies for flexible production, including applications such as pick and place, machine tending, assembly, finishing, and quality testing. Robots must be able to reliably detect and locate work pieces, reference points, and human collaborators. In logistics, one of the hot applications in collaborative robotics nowadays, manual work is predominant due to the complexity of tasks and the variation of objects in the picking and packing stations. AppliedAI solutions enter this domain more and more, especially on tasks such as order commissioning and end-of line operations, where flexibility in highly required in order to enable flexible production. 3D vision in combination with machine learning algorithms show great potential in this domain. Agile Manufacturing requires a fast adaptation of a system to new environments. It is key to reduce the amount of training data for these applications in order to achieve the KPIs.
In this workshop, approaches from research and industry are presented and enriched by end-user experiences and corresponding requirements. A major part is given to an interactive session with the attendees. The presentations and notes on the discussion about the key questions will be provided on the website and distributed among the participants. The results will be considered in the future work by the Topic Group on Perception and its contributions to the Work Programme.


Applied AI Overview: What Industry Would Like to Have
Markus Vincze Automation and Control Institute, TU Vienna

Applied AI in Agile Production
Michael Suppa, Roboception GmbH

Applied AI from the Logistics Perspective
Dr. Maximilian Beinhofer, TGW Logistics Group