This paper presents an approach to abstract capabilities of hardware components, such as a robot, and to store the necessary data for accessing that capability in Asset Administration Shells (AAS). It uses an ontology based on the Capability, Skills, and Services (CSS) model, enhanced with the possibility to describe physical relations between components. Capabilities describe abstract abilities realized through technology-neutral skills, exemplified in Docker containers using a standardized interface to communicate. The case study, bin picking, illustrates the system's application and requirements. Two user types are identified: the hardware manufacturers who can abstract and store component capabilities in an AAS, and the user who can load these shells in order to orchestrate the process program. This paper details the system's design, implementation, and evaluation, demonstrating its functionality through the case study.

The manufacturing industry is undergoing a significant transformation in the context of Industry 4.0, and production is shifting from mass products to individual products of batch size one. Moreover, the increasing complexity of components, e.g., due to additive manufacturing, makes the testing setups of components even more complex. Due to the low quantities of the components, it is not profitable to build test benches for each individual component to test a large number of different forces and torsions to ensure the needed product quality. In order to be able to test various components flexibly through different motions, we developed a concept to perform robot-based destructive component testing with industrial robots. The six degrees of freedom and the broad working range of an industrial robot make it possible to apply forces and torques to different products. Since industrial robots cannot apply the same forces and torques in all axis positions, a position must be calculated whe re the specimen can be tested. Therefore, we propose an approach for automatic specimen placement, which includes a format to map applicable forces and torques of industrial robots. Furthermore, we present an algorithmic approach to execute an automatic feasibility check for the required test motions and an automatic specimen placement using an exemplary robot-based component testing bench.

A key aspect of industry 4.0 is the transition of production to batch size one and consequently unique dimensions and structures of components for each product. Since many components are only available in small quantities it is not feasible to design expensive test benches for each of these components, however it is still important to test them to ensure the quality of each individual component. Therefore, we propose an approach for a flexibly programmable robotic test bench for destructive component testing of various components. This includes a concept for planning and execution of different test movements in a component test on robotic test benches and a unified data platform for controlling sensor-based motions as well as the recording of test data.

Industrial manufacturing is currently undergoing a transformation from mass production with inflexible production systems to individual production with adaptable cells. In order to ensure this adaptability of these systems, technologies such as plug & produce are needed, to integrate, modify and remove devices at runtime. Therefor an exact description of the system, the products and the capabilities / skills of the devices is essential as well as a network for communication between the devices. Deterministic data transmission is particularly important for distributed control systems. We propose an architecture for plug & produce mechanisms with hard real-time capable communication paths between the cyber-physical components using OPC UA PubSub over TSN and the ability to load and execute real-time critical tasks at runtime.

A key aspect of Industry 4.0 is the continuous interconnectedness of components. The standardized industrial communication protocol OPC UA offers a solution to this problem by enabling the exchange of data between the shop floor level and the inter-enterprise level. Due to the integration of the Time Sensitive Network (TSN) into OPC UA, it is now even possible to exchange information in real-time. Especially on the shop floor, there are numerous heterogeneous distributed devices from sensors to robots which must communicate with each other in real-time to achieve a distributed industrial control. Therefore, we propose an approach to combine real-time communication over TSN with OPC UA Programs to synchronize multiple distributed OPC UA Programs and exchange process data between them without losing real-time guarantees. This can be seen as the enabler of Plug-and-Produce with real-time requirements.

XCSF is a rule-based on-line learning system that makes use of local learning concepts in conjunction with gradient-based approximation techniques. It is mainly used to learn functions, or rather regression problems, by means of dividing the problem space into smaller subspaces and approximate the function values linearly therein. In this paper, we show how local interpolation can be incorporated to improve the approximation speed and thus to decrease the system error. We describe how a novel interpolation component integrates into the algorithmic structure of XCSF and thereby augments the well-established separation into the performance, discovery and reinforcement component. To underpin the validity of our approach, we present and discuss results from experiments on three test functions of different complexity, i.e. we show that by means of the proposed strategies for integrating the locally interpolated values, the overall performance of XCSF can be improved.