Martin Ruskowski

Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau | TUK · Machine Tools and Control Systems

Rising energy prices and an increasing share of volatile energy supply from renewable energy are leading to greater interest in detailed modeling of energy consumption in manufacturing. Nevertheless, energy measurements and energy load profiling at the machine level as well as the application of energy-related data for production scheduling is chal...

Compressed air is an important work medium for transfer of energy in many industrial processes. The inefficient physical processes used to produce it make compressed air one of the most expensive energy sources in supply systems. Even small leakages can over time result in high energy losses and costs if not detected and fixed timely. In addition,...

According to the guiding principles of Industry 4.0, edge computing enables the data-sovereign and near-real-time processing of data directly at the point of origin. Using these edge devices in manufacturing organization will drive the use of industrial analysis, control, and Artificial Intelligence (AI) applications close to production. The goal o...

The need to shift from a linear economy to a circular economy (CE) is not only because of the decline in raw material resources, but also because of regulations and mandates for climate and environmental protection. Digital Twins serve as enablers for this shift. However, proprietary systems and data formats create interoperability barriers between...

In the realm of Federated Learning (FL), particularly within the manufacturing sector, the strategy for selecting client weights for server aggregation is pivotal for model performance. This study investigates the comparative effectiveness of two weight selection strategies: Final Epoch Weight Selection (FEWS) and Optimal Epoch Weight Selection (OE...

Federated Learning (FL) represents a paradigm shift in the field of machine learning, offering an approach for a decentralized training of models across a multitude of devices while maintaining the privacy of local data. However, the dynamic nature of FL systems, characterized by the ongoing incorporation of new clients with potentially diverse dat...

Federated learning (FL) has emerged as a promising approach to training machine learning models across decentralized data sources while preserving data privacy, particularly in manufacturing and shared production environments. However, the presence of data heterogeneity variations in data distribution, quality, and volume across different or client...

Federated Learning (FL) has garnered significant attention in manufacturing for its robust model development and privacy-preserving capabilities. This paper contributes to research focused on the robustness of FL models in object detection, hereby presenting a comparative study with conventional techniques using a hybrid dataset for small object de...

In modern smart factories we have multiple entities that interact with one another, such as worker-assistance system, robot collaboration and their corresponding software modules. To facilitate seamless cooperation between those subsystems, we propose an Ontology-based Digital Twin that allows semantic representation of all important parts of such...

While the benefits of reconfigurable manufacturing systems (RMS) are well-known, there are still challenges to their development, including, among others, a modular software architecture that enables rapid reconfiguration without much reprogramming effort. Skill-based engineering improves software modularity and increases the reconfiguration potent...

To facilitate the shift from a Linear to a Circular Economy, it is essential to alter the manner in which we manufacture, use, and dispose of products. Interoperable information exchange among various stakeholders throughout the product lifecycle is crucial if we wish to encourage reuse, repair, remanufacturing, refurbishing, or recycling. The Digi...

Current challenges facing the manufacturing sector necessitate innovations such as modular production. This novel form of production can be dynamically aligned with market demand and provides a highly efficient, hence sustainable, production environment. A key element for flexibility and efficiency is data exchange between production machines and t...

Modular cyber-physical production systems are an important paradigm of Industry 4.0 to react flexibly to changes. The flexibility of those systems is further increased with skill-based engineering and can be used to adapt to customer requirements or to adapt manufacturing to disturbances in supply chains. Further potential for application of these...

On top of geopolitical tensions and resulting supply chain disruptions, companies are increasingly challenged to meet policy standards for the circular economy. One promising approach is extending the principles of Shared Production to the product lifecycle. This allows the implementation of a Digital Product Passport using Gaia-X principles in com...

Modular skill-based production environments provide a high flexibility to fulfil various requirements in manufacturing like the production of individualized products or the selection of alternative production chains in case of unforeseen disturbances. Within these complex and changing production environments, there is a demand for transparency on t...

This paper investigates the application of AI-based methods for characterizing waste materials in sorting processes. With the increasing use of sensors in waste sorting systems, there is an opportunity to integrate data and improve accuracy. AI methods, such as deep object detection models, have the potential to optimize waste management processes...

The primary goal of this research is to describe the scenarios, challenges, and complexities associated with object detection in industrial environments and to provide clues on how to tackle them. While object detection in production lines offers significant advantages, it also poses notable difficulties. This chapter delves into the common scenari...

This chapter aims at presenting the system architecture of a distributed production testbed embedded in an interoperable Shared Production network. The goal of the modular architecture is to enable flexible, resilient and distributed production. The presented approach illustrates how Multi-Agent Systems (MAS) can be incorporated in the manufacturin...

This paper presents a novel approach to recognize a worker’s intentions for assistive systems in manual assembly. Building on previous research, it introduces a process description for assembling products, addressing the inflexibility and limitations of previous models. This enhancement enables the modeling of assemblies for a substantial amount of...

The Digital Product Passport (DPP) is a vital enabler of the circular economy. However, creating interoperability among digital representations of products/assets and preserving privacy are core challenges for a successful implementation of the DPP. The Asset Administration Shell (AAS), by offering a consistent metamodel and application programming...

Federated learning is a collaborative machine learning approach that allows multiple parties to train a model without exchanging sensitive data. In manufacturing, where different parties may have proprietary or sensitive data that cannot be shared, this is especially useful. However, traditional federated learning approaches (as proposed by McMahan...

Increasing the flexibility of production environments and the resilience of value chains are major challenges of Industry 4.0. Multi-enterprise manufacturing networks can be a solution to create dynamic supply chains and to make product manufacturing more flexible. To enable the concept of manufacturing networks, vendor-independent and data-secure...

The rising variability of production plants is one of the central challenges in industry, independent of the company size. Although, wireless communication has become an integral part of modern industrial systems with the introduction of the fifth generation (5G) of wireless technology, flexible and modular manufacturing systems have particular dem...

Changing markets, individualized products, and volatile supply chains require a high degree of flexibility, especially in production and manufacturing. A promising approach to reconfigurable and flexible manufacturing is the use of machine-level skills. This paper presents a simple architecture for skill-based machining of individualized milled par...

Um Produkte zwischen Produktionsmodulen und autonomen mobilen Roboter (AMR) austauschen zu können werden aktuell komplexe Individuallösungen entwickelt. Diese sind auf einer zentralen Steuerung mit statisch vorprogrammierten Abläufen und einer physischen Positionierung des AMR umgesetzt. Diese Lösungen sind häufig nicht auf andere Implementierungen...

The use of distributed optimization in machine learning can be motivated either by the resulting preservation of privacy or the increase in computational efficiency. On the one hand, training data might be stored across multiple devices. Training a global model within a network where each node only has access to its confidential data requires the u...

Federated learning (FL) has emerged as a promising approach for training machine learning models on decentralized data without compromising data privacy. In this paper, we propose a FL algorithm for object detection in quality inspection tasks using YOLOv5 as the object detection algorithm and Federated Averaging (FedAvg) as the FL algorithm. We ap...

Federated learning (FL) has gained significant traction as a privacy-preserving algorithm, but the underlying resembles of federated learning algorithm like Federated averaging (FED Avg) or Federated SGD (FED SGD) to ensemble learning algorithms has not been fully explored. The purpose of this paper is to examine the application of FL to object det...

In diesem Beitrag wird ein Ansatz für die Umsetzung der Shared Production basierend auf Konzepten der Plattform Industrie 4.0 und Gaia-X vorgestellt. Das Ziel der Shared Production ist der Aufbau eines Fertigungsnetzwerkes mit einer automatisierten Einbindung von externen Produktionskapazitäten. Dies ermöglicht die Ad-hoc-Erstellung von Lieferkette...

Modular production systems enable resilient production processes through decoupled production processes. On the way to implementing flexible and adaptable production systems, information support plays a decisive role. Only the use of intelligent and structured information processing across previous system boundaries and areas enables the coordinati...

Cyber-Physical Production Modules (CPPMs) must be described by vendor-independent and machine-readable standardized information models. Standards make CPPMs adaptable and interchangeable at different company levels to enable flexible production. We present an OPC UA information model for CPPMs based on the relevant OPC UA Companion Specifications....

In this paper, two new algorithms for dual decomposition-based distributed optimization are presented. Both algorithms rely on the quadratic approximation of the dual function of the primal optimization problem. The dual variables are updated in each iteration through a maximization of the approximated dual function. The first algorithm approximate...

Im Auftrag des Bundesministeriums für Wirtschaft und Klimaschutz haben DIN und DKE im Januar 2022 die Arbeiten an der zweiten Ausgabe der Deutschen Normungsroadmap Künstliche Intelligenz gestartet. In einem breiten Beteiligungsprozess und unter Mitwirkung von mehr als 570 Fachleuten aus Wirtschaft, Wissenschaft, öffentlicher Hand und Zivilgesellsch...

Im Bereich der modularen und flexiblen Produktion stellen konventionelle Ansätze für die Maschinensicherheit (Safety) Hürden in der Produktivität und Wirtschaftlichkeit dar. In diesem Artikel wird ein Konzept mit Fokus einer digitalen Produktionsplanung und -durchführung für Betreiber diskreter Produktionen entwickelt, welches die Sicherheits-Probl...

To improve the fault tolerance of production systems against unforeseen events, modern production systems must be able to react more autonomously with suitable solutions to reach their goals. Such events within the company, e.g., resource breakdowns, require the possibility of internal replanning of the production schedule. Multi-Agent Systems (MAS...

The Digital Product Passport (DPP) is a concept to collect and share product-related information along a product's lifecycle. The aim is to provide all stakeholders during the product lifecycle with the information they need such that a successful Circular Economy can be implemented. At the moment, several varieties of DPPs are being developed, mos...

The flexibilization and increase of new production concepts such as matrix manufacturing with the help of autonomous logistics robots (AGVs) pose new challenges to production scheduling. To solve these flexible job shop scheduling problems (FJSSP) for arbitrary production arrangements, a concept for a multi-agent system based on Deep Reinforcement...

A skill-based engineering approach uses the concept of skill to abstract machine-specific functionality with a generic interface and common behavior. A skill is treated as a “control level service” and is used to build service-oriented architectures. Though the question of robust and flexible coordination of skills is still open. Current approaches...

A vast amount of data is created every minute, both in the private sector and industry. Whereas it is often easy to get hold of data in the private entertainment sector, in the industrial production environment it is much more difficult due to laws, preservation of intellectual property, and other factors. However, most machine learning methods req...

Das Whitepaper gibt einen Einblick in die Vision von Production Level 4 und die zugrunde liegenden wissenschaftlichen Arbeiten, die Technologien und notwendigen Architekturen. Zentraler Aspekt einer nachhaltigen hoch verfügbaren Produktion ist die Verlässlichkeit der Prozesse, Maschinen und Abläufe. Der Einsatz von Methoden der Künstlichen Intelli...

Volatile markets with strongly fluctuating customer demand and uncertainties in the global supply chain challenge manufacturing companies with increasing complexity. A flexible, reconfigurable production system like the concept of Modular Production by Kern is one way to meet these challenges and at the same time to produce more efficiently. Such p...

Es gibt mehrere vorgeschlagene Architekturen für das Edge Computing, aber es gibt bislang keine von der Community oder der Industrie akzeptierten Standards. Außerdem gibt es keine gemeinsame Vereinbarung darüber, wie die Edge Computing-Architektur physisch aussieht. In diesem Artikel wird die Industrial Edge Cloud beschrieben, erklärt, wie eine Ind...

The Circular Economy approach aims to close the loop of materials and to reduce waste. However, relevant product data for the optimization and management of circular approaches are often missing. Stakeholders typically lack key information: Recyclers do not know which materials/compounds to expect, producers do not know enough about the recyclabili...

Job-shop scheduling problems are important in the industrial context to achieve high machine utilization. Heuristics offer a possibility to solve these problems with moderate computational effort. However, they might be associated with a high development effort and generalization to other tasks is difficult. We use a reinforcement learning approach...

A flexible operation of multiple robotic manipulators operating in a dynamic environment requires online trajectory planning to ensure collision-free trajectories. In this work, we propose a real-time capable motion control algorithm, based on nonlinear model predictive control, which accounts for static and dynamic obstacles. The proposed algorith...

Increasing demand for customized products in the wake of the 4th Industrial Revolution is placing ever increasing demands on the flexibility of manufacturing systems. Furthermore, the increasing usage of automated guided vehicles (AGV) adds another layer of flexibility and also complexity to the overall production system. The resulting Flexible Job...

Runge–Kutta neural networks (RKNN) bridge the gap between continuous-time advantages and the discrete-time nature of any digital controller. RKNN defines a neural network in a continuous-time setting and explicitly discretises it by a variable sample time Runge–Kutta method. As a result, a-priori model knowledge such as the well-known continuous-ti...

Functionally described capabilities play an important role in the virtualization of manufacturing and the resource-specific realization with the skill-based approach. The importance of this modeling can be seen in a formalization of production capabilities for a holistic usage in a workflow from product orders towards manufacturing on the shop floo...

Detecting small objects in video streams of head-worn augmented reality devices in near real-time is a huge challenge: training data is typically scarce, the input video stream can be of limited quality, and small objects are notoriously hard to detect. In industrial scenarios, however, it is often possible to leverage contextual knowledge for the...