Plug, Plan, and Produce for Reconfiguration of Modular and Robot-Based Production Systems

Smart Factories, collaborative robots, and cyber-physical systems revolutionize the production processes and reshape the industrial landscape initiated by Industry 4.0 developments. Within this changing environment, the role of the human worker shifts from the manual production of goods to the supervision of automated processes. Workers have to master new skills such as programming collaborative robots, arrange its workspace, and put the workcell into operation.<br /><br /> As these are new tasks, I introduce a Modular Production System that assists the nonprofessional robotic users with helpful, context-sensitive instructions. To this end, I contribute a methodology I coin “Plug, Plan, and Produce“ that realizes a bidirectional commissioning process with modular components for flexible adjustment of manufacturing processes. An extended Plug-and-Produce concept enables the physical connection of components, wherein the operator explores the workspace utilizing the collaborative robot as an advanced localization tool. During a dedicated planning phase, an interactive user interface facilitates an embedded kinematic simulation to visualize robot motions, to represent the accurate state of workspace, and to compute dynamically collision-free trajectories. Components are specified by a Self-Description to ensure a seamless data-driven integration process providing a distinct function description, an entire representation of robotic task frames, and CAD-models for visualization and collision avoidance purposes of the simulation.<br /><br /> Targeting the customized and small-batched production with frequent reconfiguration demands, a Skill-Based Architecture allows the configuration, composition, and parameterization of executable workflows to control the robot and production system. Further, I contribute an Interactive Workspace Assistant to generate, simulate, and evaluate component layouts enhancing the assistance of implicit robot programming and commissioning processes.<br /><br /> I validated the presented methodology against a real-world setup evaluating the robotic system in an extensive user study where participants performed a flexible workcell setup with tasks of increasing complexity. The results support the hypothesis that the interactive assistance supports the commissioning process of users conducting a previously unknown robotic tasks in an industrial assembly scenario.

Level thesis:degree_level

thesis.doctoral

Grantor dc:publisher

Universität Bielefeld

Year

2022

Author dc:creator

• Wojtynek, Michael

Repository record source_url

https://pub.uni-bielefeld.de/record/2968429

OAI identifier oai:identifier

oai:pub.uni-bielefeld.de:2968429