Veröffentlichungen

In diesem Beitrag wird die systematische Entwicklung einer Schraubeinheit für Industrieroboter vorgestellt, mit derer Hilfe ein Roboter die Demontage von Elektromotoren vornehmen kann. Aufgrund der Anwendung im Remanufacturing, haben die Elektromotoren bereits einen Lebenszyklus durchlaufen und zeichnen sich durch ungewisse Produktzustände aus. Der Schraubeffektor ist unabhängig davon im Stande, verschiedene Schrauben zu lösen und dabei Messwerte aufzunehmen.

Cognitive abilities allow robots to learn and reason from their environment. The gained knowledge can then be incorporated into the robot’s actions which in turn affect the environment. Therefore, a cognitive robot is no longer a static system that performs actions based on a pre-defined set of rules but a complex entity that dynamically adjusts over time. With this, challenges arise for production systems that need to observe and ideally anticipate the cognitive robot’s behavior. Often, digital twins are employed to test and optimize production control systems. This paper presents a generic approach to characterize, model and simulate learning processes and formalized knowledge in hybrid production systems assuming different station types with learning effects. Thereby, quantitative and qualitative learning processes are mapped including knowledge sharing and transfer across entities. A modular and parameterizable design enables the adjustment to different use cases. Eventually, the model is instantiated as a digital twin of a real production system for product disassembly employing cognitive-autonomous robots among human operators and rigidly automated machines. The model shows great potential to be integrated into test beds for planning and control systems of cognitive factories.

Production systems must be able to quickly adapt to changing requirements. Especially in the field of remanufacturing, the uncertainty in the state of the incoming products is very high. Several adaptation mechanisms can be applied leading to agile and changeable production systems. Among these, adapting the degree of automation with respect to changeover times and high investment costs is one of the most challenging mechanisms. However, not only long-term changes, but also short-term adaptations can lead to enormous potentials, e.g. when night shifts can be supported by robots and thus higher labor costs and unfavorable working conditions at night can be avoided. These changes in the degree of automation on an operational level are referred to as fluid automation, which will be defined in this paper. The mechanisms of fluid automation are presented together with a case study showing its application on a disassembly station for electrical drives.

An increase in the sales number of battery electric vehicles within the last year can be recorded. At the end-of-life these vehicles require a reliable disassembly for recycling or remanufacturing. On the one hand, drivetrain components of those vehicles contain valuable resources and thus are mainly relevant for recycling or remanufacturing. On the other hand, the automated disassembly of especially electric motors and Li-ion battery systems encloses major challenges. Especially the high number of variants and the unknown specifications and conditions of the components are challenging points for the disassembly system. Conventional automated disassembly systems provide limited flexibility and adaptability for the disassembly of these products. Within this contribution two robot-based flexible disassembly systems are systematically derived for Li-ion battery modules and supplementary electric motors. Both products are analysed and the product-specific challenges and requirements are identified. The state of the art regarding flexible disassembly systems is captured using the methodology of a morphological box. Four subsystems are identified: Kinematic, Tools, Workpiece fixation, Safety system. Based on the results, concepts for disassembly systems for both Li-ion battery modules and supplementary electric motors are developed and presented in detail. Especially the structure and functionality of both systems are explained. This is followed by an assessment of the approaches and an identification of limitations as well as possible optimization potentials.

Das Remanufacturing, bisher geprägt durch manuelle und kostenintensive Prozesse, ist ein entscheidender Schritt auf dem Weg zu einer ressourcenschonenden Kreislaufwirtschaft. Industrie und Forschung sind sich einig, dass der Einzug von Industrie 4.0 Technologien den Schlüssel zu einer Entwicklung automatisierter und wirtschaftlicher Remanufacturing-Systeme darstellt. Basierend auf einer systematischen Literaturrecherche widmet sich dieser Beitrag der Analyse vielversprechender Industrie 4.0-Ansätze mit dem Fokus auf den übergeordneten Gesamtprozess sowie den Teilprozessen der Demontage und der Inspektion. Die Ergebnisse legen nahe, dass es an zusätzlichem Wissen, Erfahrung und Forschung bei der Entwicklung und realen Demonstration der Ansätze und deren Übertragbarkeit auf breitere Anwendungsfelder bedarf.

Inspektionsprozesse werden im Remanufacturing auch heute noch vorwiegend manuell durchgeführt, da die Einschätzung des Qualitätszustands von rückläufigen Gebrauchtprodukten komplex und damit schwer zu automatisieren ist. Dies ist darauf zurückzuführen, dass Abnutzungsgrade, Deformationen und Schädigungen eine individuelle Bewertung des Gebrauchtprodukts nach sich ziehen und somit schwer standardisierbar sind. In diesem Beitrag werden die Anforderungen an ein System für die Bewältigung der Herausforderung der automatisierten Inspektion im Remanufacturing abgeleitet. Darauf aufbauend wird das Konzept einer Befundungsstation, welches diese Anforderungen erfüllt, präsentiert und Anwendungsfälle im Rahmen des von der CarlZeiss-Stiftung geförderten Forschungsprojekts „AgiProbot - Agiles Produktionssystem mittels mobiler, lernender Roboter mit Multisensorik bei ungewissen Produktspezifikationen“ vorgestellt.

Remanufacturing, previously characterised by manual and cost-intensive processes, is a decisive step towards a resource-conserving circular economy. Uncertain product states, inconsistent quality, and fluctuating availability of end-of-life products not only pose major challenges for the automation of remanufacturing, but also for intralogistics, which has hardly been considered in the literature to date. This paper gives a concept overview on an ontology-based autonomous intralogistics system embedded in the fluid automation framework, describes and illustrates the main cyber-physical components and shows exemplary workflows. The presented concept is currently implemented as part of the AgiProbot research project.

Eine Automatisierung der Inspektion im Remanufacturing bietet die Möglichkeit, Kostenpotenziale zu erschließen. Dies lässt sich mittels robotergeführter optischer Messsysteme erreichen. In diesem Beitrag werden bestehende Ansätze vorgestellt und Herausforderungen für View-Planning-Ansätze diskutiert, welche sich aus den Besonderheiten des Remanufacturing heraus ergeben. Gleichzeitig werden Lösungsansätze für diese Problemstellungen an einem beispielhaften Anwendungsfall aufgezeigt.

Robot based remanufacturing of valuable products is commonly perceived as promising field in future in terms of an efficient and globally competitive economy. Additionally, it plays an important role with regard to resource-efficient manufacturing. The associated processes however, require a reliable non-destructive disassembly. For these disassembly processes, there is special robot periphery essential to enable the tasks physically. Unlike manufacturing, within remanufacturing there are End-of-Life (EoL) products utilized. The specifications and conditions are often uncertain and varying. Consequently the robot system and especially the periphery needs to adapt to the used product, based on an initial examination and classification of the part. State of the art approaches provide limited flexibility and adaptability to the disassembly of electric motors used in automotive industry. Especially the geometrical shape is a limiting factor for using state of the art periphery for remanufacturing. Within this contribution a new kind of flexible clamping device for the disassembly of EoL electrical motors is presented. The robot periphery is systematically developed

Remanufacturing includes disassembly and reassembly of used products to save natural resources and reduce emissions. While assembly is widely understood in the field of operations management, disassembly is a rather new problem in production planning and control. The latter faces the challenge of high uncertainty of type, quantity and quality conditions of returned products, leading to high volatility in remanufacturing production systems. Traditionally, disassembly is a manual labor-intensive production step that, thanks to advances in robotics and artificial intelligence, starts to be automated with autonomous workstations. Due to the diverging material flow, the application of production systems with loosely linked stations is particularly suitable and, owing to the risk of condition induced operational failures, the rise of hybrid disassembly systems that combine manual and autonomous workstations can be expected. In contrast to traditional workstations, autonomous workstations can expand their capabilities but suffer from unknown failure rates. For such adverse conditions a condition-based control for hybrid disassembly systems, based on reinforcement learning, alongside a comprehensive modeling approach is presented in this work. The method is applied to a real-world production system. By comparison with a heuristic control approach, the potential of the RL approach can be proven simulatively using two different test cases.

Remanufacturing is commonly perceived as a promising field for future challenges such as resource efficient production. For an economic operation of remanufacturing facilities, an automation of the currently manual labor is mandatory. Thus, the automation plays a vital role in order to realize high rates of re-utilization and therefore a significant reduction of waste. Screw connections allow for non-destructive dismantling and are commonly used connection elements. Especially the automation of the disassembly step is a key element as products from the field are of unknown specification upon feeding to the remanufacturing line due to alterations during their life cycles. State of the art solutions for automated disassembly lack flexibility to adapt to different products and product conditions. This contribution presents a highly flexible approach for the localization and classification of screws in electric motors. The presented system utilizes a tool equipped industrial robot with an integrated eye-in-hand vision system and an industrial computer. The system is able to locate and classify six different types of screw heads of varying sizes using machine learning approaches in order to adapt the robot’s end-effector. Because of the presented hardware concept the system depends upon a minimum of constraints concerning the presentation of objects. This paper compares different network architectures and peripheral settings and presents the most suitable solution to the use case. A dataset consisting of six classes of different screw heads was created to train neural networks to detect screws in an experimental set-up consisting of metal blocks holding different screws of diverse types and conditions. Results are validated on two different electric motors from the automotive sector being processed on an automated disassembly line.

Agile production systems combine a high degree of flexibility and adaptability. These qualities are particularly crucial in an environment with high uncertainty, for example in the context of remanufacturing. Remanufacturing describes the industrial process of reconditioning used parts so that they regain comparable technical properties as new parts. Due to the scarcity of resources and regulatory requirements, the importance of remanufacturing is increasing. Due to the unpredictable component properties, automation plays a subordinate role in remanufacturing. The authors present a concept how automated disassembly can be achieved even for components of uncertain specifications by using artificial intelligence. For the autonomous development of disassembly capabilities, digital twins are used as learning environments. On the other hand, skills and problem-solving strategies are identified and abstracted from human observation. To achieve an efficient disassembly system, a modular station concept is applied, both on the technical and on the information technology level.

Competitive remanufacturing of used products with uncertain conditions requires a high degree of flexibility and responsiveness. This article describes an integrated control architecture for a modular, agile disassembly system with autonomous product inspection and learning production resources. The approach includes a material flow control and vertically-integrated sub-architectures to control the station and intralogistics operations.

Recognizing human actions is a vital task for a humanoid robot, especially in domains like programming by demonstration. Previous approaches on action recognition primarily focused on the overall prevalent action being executed, but we argue that bimanual human motion cannot always be described sufficiently with a single action label. We present a system for framewise action classification and segmentation in bimanual human demonstrations. The system extracts symbolic spatial object relations from raw RGB-D video data captured from the robot's point of view in order to build graph-based scene representations. To learn object-action relations, a graph network classifier is trained using these representations together with ground truth action labels to predict the action executed by each hand. We evaluated the proposed classifier on a new RGB-D video dataset showing daily action sequences focusing on bimanual manipulation actions. It consists of 6 subjects performing 9 tasks with 10 repetitions each, which leads to 540 video recordings with 2 hours and 18 minutes total playtime and per-hand ground truth action labels for each frame. We show that the classifier is able to reliably identify (action classification macro F1-score of 0.86) the true executed action of each hand within its top 3 predictions on a frame-by-frame basis without prior temporal action segmentation.

We introduce a multi-functional robotic gripperequipped with a set of actions required for disassembly ofelectromechanical devices. The gripper consists of a robot armwith 5 degrees of freedom (DoF) for manipulation and a jawgripper with a 1-DoF rotation joint and a 1-DoF closing joint.The system enables manipulation in 7 DoF and offers the abilityto reposition objects in hand and to perform tasks that usuallyrequire bimanual systems. The sensor system of the gripperincludes relative and absolute joint encoders, force and pressuresensors to provide feedback about interaction forces, a tool-mounted camera for screw detection and precise placement ofthe tool tip using image-based visual servoing. We present adata-driven method for estimating joint torques based on theoutput voltage and motor speed. Further, we provide methodsfor teaching disassembly actions based on human demonstra-tion, their representation as movement primitives and executionbased on sensory feedback. We provide quantitative resultsregarding positioning and torque estimation accuracy, disassem-bly success rate and qualitative results regarding the successfuldisassembly of hard disc drives.