Michael Weyrich

Yuchen Xia, Manthan Shenoy, Nasser Jazdi et al.

In this paper, we present a novel framework that combines large language models (LLMs), digital twins and industrial automation system to enable intelligent planning and control of production processes. We retrofit the automation system for a modular production facility and create executable control interfaces of fine-granular functionalities and coarse-granular skills. Low-level functionalities are executed by automation components, and high-level skills are performed by automation modules. Subsequently, a digital twin system is developed, registering these interfaces and containing additional descriptive information about the production system. Based on the retrofitted automation system and the created digital twins, LLM-agents are designed to interpret descriptive information in the digital twins and control the physical system through service interfaces. These LLM-agents serve as intelligent agents on different levels within an automation system, enabling autonomous planning and control of flexible production. Given a task instruction as input, the LLM-agents orchestrate a sequence of atomic functionalities and skills to accomplish the task. We demonstrate how our implemented prototype can handle un-predefined tasks, plan a production process, and execute the operations. This research highlights the potential of integrating LLMs into industrial automation systems in the context of smart factory for more agile, flexible, and adaptive production processes, while it also underscores the critical insights and limitations for future work. Demos at: https://github.com/YuchenXia/GPT4IndustrialAutomation

Yuchen Xia, Nasser Jazdi, Jize Zhang et al.

Traditional industrial automation systems require specialized expertise to operate and complex reprogramming to adapt to new processes. Large language models offer the intelligence to make them more flexible and easier to use. However, LLMs' application in industrial settings is underexplored. This paper introduces a framework for integrating LLMs to achieve end-to-end control of industrial automation systems. At the core of the framework are an agent system designed for industrial tasks, a structured prompting method, and an event-driven information modeling mechanism that provides real-time data for LLM inference. The framework supplies LLMs with real-time events on different context semantic levels, allowing them to interpret the information, generate production plans, and control operations on the automation system. It also supports structured dataset creation for fine-tuning on this downstream application of LLMs. Our contribution includes a formal system design, proof-of-concept implementation, and a method for generating task-specific datasets for LLM fine-tuning and testing. This approach enables a more adaptive automation system that can respond to spontaneous events, while allowing easier operation and configuration through natural language for more intuitive human-machine interaction. We provide demo videos and detailed data on GitHub: https://github.com/YuchenXia/LLM4IAS.

Yuchen Xia, Jize Zhang, Nasser Jazdi et al.

This paper introduces a novel approach to integrating large language model (LLM) agents into automated production systems, aimed at enhancing task automation and flexibility. We organize production operations within a hierarchical framework based on the automation pyramid. Atomic operation functionalities are modeled as microservices, which are executed through interface invocation within a dedicated digital twin system. This allows for a scalable and flexible foundation for orchestrating production processes. In this digital twin system, low-level, hardware-specific data is semantically enriched and made interpretable for LLMs for production planning and control tasks. Large language model agents are systematically prompted to interpret these production-specific data and knowledge. Upon receiving a user request or identifying a triggering event, the LLM agents generate a process plan. This plan is then decomposed into a series of atomic operations, executed as microservices within the real-world automation system. We implement this overall approach on an automated modular production facility at our laboratory, demonstrating how the LLMs can handle production planning and control tasks through a concrete case study. This results in an intuitive production facility with higher levels of task automation and flexibility. Finally, we reveal the several limitations in realizing the full potential of the large language models in autonomous systems and point out promising benefits. Demos of this series of ongoing research series can be accessed at: https://github.com/YuchenXia/GPT4IndustrialAutomation

Benjamin Maschler, Angel Iliev, Thi Thu Huong Pham et al.

Real-life industrial use cases for machine learning oftentimes involve heterogeneous and dynamic assets, processes and data, resulting in a need to continuously adapt the learning algorithm accordingly. Industrial transfer learning offers to lower the effort of such adaptation by allowing the utilization of previously acquired knowledge in solving new (variants of) tasks. Being data-driven methods, the development of industrial transfer learning algorithms naturally requires appropriate datasets for training. However, open-source datasets suitable for transfer learning training, i.e. spanning different assets, processes and data (variants), are rare. With the Stuttgart Open Relay Degradation Dataset (SOReDD) we want to offer such a dataset. It provides data on the degradation of different electromechanical relays under different operating conditions, allowing for a large number of different transfer scenarios. Although such relays themselves are usually inexpensive standard components, their failure often leads to the failure of a machine as a whole due to their role as the central power switching element of a machine. The main cost factor in the event of a relay defect is therefore not the relay itself, but the reduced machine availability. It is therefore desirable to predict relay degradation as accurately as possible for specific applications in order to be able to replace relays in good time and avoid unplanned machine downtimes. Nevertheless, data-driven failure prediction for electromechanical relays faces the challenge that relay degradation behavior is highly dependent on the operating conditions, high-resolution measurement data on relay degradation behavior is only collected in rare cases, and such data can then only cover a fraction of the possible operating environments. Relays are thus representative of many other central standard components in automation technology.

Timo Müller, Benjamin Maschler, Daniel Dittler et al.

Many decision-making approaches rely on the exploration of solution spaces with regards to specified criteria. However, in complex environments, brute-force exploration strategies are usually not feasible. As an alternative, we propose the combination of an exploration task's vertical sub-division into layers representing different sequentially interdependent sub-problems of the paramount problem and a horizontal sub-division into self-sustained solution sub-spaces. In this paper, we present a universal methodology for the intelligent exploration of solution spaces and derive a use-case specific example from the field of reconfiguration management in industry 4.0.

Benjamin Maschler, Tim Knodel, Michael Weyrich

Industrial transfer learning increases the adaptability of deep learning algorithms towards heterogenous and dynamic industrial use cases without high manual efforts. The appropriate selection of what to transfer can vastly improve a transfer's results. In this paper, a transfer case selection based upon clustering is presented. Founded on a survey of clustering algorithms, the BIRCH algorithm is selected for this purpose. It is evaluated on an industrial time series dataset from a discrete manufacturing scenario. Results underline the approaches' applicability caused by its results' reproducibility and practical indifference to sequence, size and dimensionality of (sub-)datasets to be clustered sequentially.

Matthias Weiß, Falk Dettinger, Elias Detrois et al.

Real-world implementations of connected vehicle functions are spreading steadily, yet operating these functions reliably remains challenging due to their distributed nature and the complexity of the underlying cloud, edge, and networking infrastructure. Quick diagnosis of problems and understanding the error chains that lead to failures is essential for reducing downtime. However, diagnosing these systems is still largely performed manually, as automated analysis techniques are predominantly data-driven and struggle with hidden relationships and the integration of context information. This paper addresses this gap by introducing a multimodal approach that integrates human feedback and system-specific information into the causal analysis process. Reinforcement Learning from Human Feedback is employed to continuously train a causality mining model while incorporating expert knowledge. Additional modules leverage distributed tracing data to prune false-positive causal links and enable the injection of domain-specific relationships to further refine the causal graph.Evaluation is performed using an automated valet parking application operated in a connected vehicle test field. Results demonstrate a significant increase in precision from 14\% to 100\% for the detection of causal edges and improved system interpretability compared to purely data-driven approaches, highlighting the potential for system operators in the connected vehicle domain.

Yuchen Xia, Zhewen Xiao, Nasser Jazdi et al.

This research introduces a novel approach for achieving semantic interoperability in digital twins and assisting the creation of Asset Administration Shell (AAS) as digital twin model within the context of Industry 4.0. The foundational idea of our research is that the communication based on semantics and the generation of meaningful textual data are directly linked, and we posit that these processes are equivalent if the exchanged information can be serialized in text form. Based on this, we construct a "semantic node" data structure in our research to capture the semantic essence of textual data. Then, a system powered by large language models is designed and implemented to process the "semantic node" and generate standardized digital twin models from raw textual data collected from datasheets describing technical assets. Our evaluation demonstrates an effective generation rate of 62-79%, indicating a substantial proportion of the information from the source text can be translated error-free to the target digital twin instance model with the generative capability of large language models. This result has a direct application in the context of Industry 4.0, and the designed system is implemented as a data model generation tool for reducing the manual effort in creating AAS model. In our evaluation, a comparative analysis of different LLMs and an in-depth ablation study of Retrieval-Augmented Generation (RAG) mechanisms provide insights into the effectiveness of LLM systems for interpreting technical concepts and translating data. Our findings emphasize LLMs' capability to automate AAS instance creation and contribute to the broader field of semantic interoperability for digital twins in industrial applications. The prototype implementation and evaluation results are presented on our GitHub Repository: https://github.com/YuchenXia/AASbyLLM.

Falk Dettinger, Matthias Weiß, Daniel Baumann et al.

Cloud-based offloading helps address energy consumption and performance challenges in executing resource-intensive vehicle algorithms. Utilizing 5G, with its low latency and high bandwidth, enables seamless vehicle-to-cloud integration. Currently, only non-standalone 5G is publicly available, and real-world applications remain underexplored compared to theoretical studies. This paper evaluates 5G non-standalone networks for cloud execution of vehicle functions, focusing on latency, Round Trip Time, and packet delivery. Tests used two AI-based algorithms -- emotion recognition and object recognition -- along an 8.8 km route in Baden-Württemberg, Germany, encompassing urban, rural, and forested areas. Two platforms were analyzed: a cloudlet in Frankfurt and a cloud in Mannheim, employing various deployment strategies like conventional applications and containerized and container-orchestrated setups. Key findings highlight an average signal quality of 84 %, with no connectivity interruptions despite minor drops in built-up areas. Packet analysis revealed a Packet Error Rate below 0.1 % for both algorithms. Transfer times varied significantly depending on the geographical location and the backend servers' network connections, while processing times were mainly influenced by the computation hardware in use. Additionally, cloud offloading seems only be a suitable option, when a round trip time of more than 150 ms is possible.

Matthias Weiß, Philipp Marks, Benjamin Maschler et al.

Nowadays, formal methods are used in various areas for the verification of programs or for code generation from models in order to increase the quality of software and to reduce costs. However, there are still fields in which formal methods have not been widely adopted, despite the large set of possible benefits offered. This is the case for the area of programmable logic controllers (PLC). This article aims to evaluate the potential of formal methods in the context of PLC development. For this purpose, the general concepts of formal methods are first introduced and then transferred to the PLC area, resulting in an engineering-oriented description of the technology that is based on common concepts from PLC development. Based on this description, PLC professionals with varying degrees of experience were interviewed for their perspective on the topic and to identify possible use cases within the PLC domain. The survey results indicate the technology's high potential in the PLC area, either as a tool to directly support the developer or as a key element within a model-based systems engineering toolchain. The evaluation of the survey results is performed with the aid of a demo application that communicates with the Totally Integrated Automation Portal from Siemens and generates programs via Fastsynth, a model-based open source code generator. Benchmarks based on an industry-related PLC project show satisfactory synthesis times and a successful integration into the workflow of a PLC developer.

Thorben Schey, Khaled Karoonlatifi, Michael Weyrich et al.

This paper introduces a novel closed-loop testing methodology for efficient linearity testing of high-resolution Successive Approximation Register (SAR) Analog-to-Digital Converters (ADCs). Existing test strategies, including histogram-based approaches, sine wave testing, and model-driven reconstruction, often rely on dense data acquisition followed by offline post-processing, which increases overall test time and complexity. To overcome these limitations, we propose an adaptive approach that utilizes an iterative behavioral model refined by an Extended Kalman Filter (EKF) in real time, enabling direct estimation of capacitor mismatch parameters that determine INL behavior. Our algorithm dynamically selects measurement points based on current model uncertainty, maximizing information gain with respect to parameter confidence and narrowing sampling intervals as estimation progresses. By providing immediate feedback and adaptive targeting, the proposed method eliminates the need for large-scale data collection and post-measurement analysis. Experimental results demonstrate substantial reductions in total test time and computational overhead, highlighting the method's suitability for integration in production environments.

Thorben Schey, Khaled Karoonlatifi, Michael Weyrich et al.

This paper builds on our Uncertainty-Guided Live Measurement Sequencing (UGLMS) method. UGLMS is a closed-loop test strategy that adaptively selects SAR ADC code edges based on model uncertainty and refines a behavioral mismatch model in real time via an Extended Kalman Filter (EKF), eliminating full-range sweeps and offline post-processing. We introduce an enhanced UGLMS that delivers significantly faster test runtimes while maintaining estimation accuracy. First, a rank-1 EKF update replaces costly matrix inversions with efficient vector operations, and a measurement-aligned covariance-inflation strategy accelerates convergence under unexpected innovations. Second, we extend the static mismatch model with a low-order carrier polynomial to capture systematic nonlinearities beyond pure capacitor mismatch. Third, a trace-based termination adapts test length to convergence, preventing premature stops and redundant iterations. Simulations show the enhanced UGLMS reconstructs full Integral- and Differential-Non-Linearity (INL/DNL) in just 36 ms for 16-bit and under 70 ms for 18-bit ADCs (120 ms with the polynomial extension). Combining the faster convergence from covariance inflation with reduced per-iteration runtime from the rank-1 EKF update, the method reaches equal accuracy 8x faster for 16-bit ADCs. These improvements enable real-time, production-ready SAR ADC linearity testing.

Falk Dettinger, Matthias Weiß, Baran Can Gül et al.

Software Defined Vehicles face an increasing computational gap as advanced algorithms and frequent software updates demand more processing power while onboard hardware remains static throughout a vehicle's 10+ year lifespan. This mismatch threatens the performance of safety-critical functions including advanced driver-assistance systems and real-time perception tasks. We propose a novel four-layer computation offloading pipeline that dynamically distributes vehicular functions to cloud and edge resources while meeting strict Round Trip Time constraints. Our key contribution is an enhanced Particle Swarm Optimization algorithm that integrates distance- and direction-based penalties with functional requirements to optimize edge server selection for mobile vehicles. Evaluation using a Kubernetes-based cloud infrastructure with realistic vehicular mobility patterns demonstrates that our approach reduces average response time compared to conventional Brute-Force methods while maintaining the success rate for latency-critical tasks. The modified Particle Swarm Optimization algorithm achieves an average execution time of 26 ms across ten servers and tasks on Central Processing Unit, and 550ms across 15 servers with 1000 tasks on Graphics Processing Unit. These results confirm the pipeline's effectiveness in bridging the computational gap for next-generation Software Defined Vehicles (SDV).

Baran Can Gül, Stefanos Tziampazis, Nasser Jazdi et al.

As Federated Learning (FL) expands to larger and more distributed environments, consistency in training is challenged by network-induced delays, clock unsynchronicity, and variability in client updates. This combination of factors may contribute to misaligned contributions that undermine model reliability and convergence. Existing methods like staleness-aware aggregation and model versioning address lagging updates heuristically, yet lack mechanisms to quantify staleness, especially in latency-sensitive and cross-regional deployments. In light of these considerations, we introduce \emph{SyncFed}, a time-aware FL framework that employs explicit synchronization and timestamping to establish a common temporal reference across the system. Staleness is quantified numerically based on exchanged timestamps under the Network Time Protocol (NTP), enabling the server to reason about the relative freshness of client updates and apply temporally informed weighting during aggregation. Our empirical evaluation on a geographically distributed testbed shows that, under \emph{SyncFed}, the global model evolves within a stable temporal context, resulting in improved accuracy and information freshness compared to round-based baselines devoid of temporal semantics.

Baran Can Gül, Suraksha Nadig, Stefanos Tziampazis et al.

In-vehicle emotion recognition underpins adaptive driver-assistance systems and, ultimately, occupant safety. However, practical deployment is hindered by (i) modality fragility - poor lighting and occlusions degrade vision-based methods; (ii) physiological variability - heart-rate and skin-conductance patterns differ across individuals; and (iii) privacy risk - centralized training requires transmission of sensitive data. To address these challenges, we present FedMultiEmo, a privacy-preserving framework that fuses two complementary modalities at the decision level: visual features extracted by a Convolutional Neural Network from facial images, and physiological cues (heart rate, electrodermal activity, and skin temperature) classified by a Random Forest. FedMultiEmo builds on three key elements: (1) a multimodal federated learning pipeline with majority-vote fusion, (2) an end-to-end edge-to-cloud prototype on Raspberry Pi clients and a Flower server, and (3) a personalized Federated Averaging scheme that weights client updates by local data volume. Evaluated on FER2013 and a custom physiological dataset, the federated Convolutional Neural Network attains 77% accuracy, the Random Forest 74%, and their fusion 87%, matching a centralized baseline while keeping all raw data local. The developed system converges in 18 rounds, with an average round time of 120 seconds and a per-client memory footprint below 200 MB. These results indicate that FedMultiEmo offers a practical approach to real-time, privacy-aware emotion recognition in automotive settings.

Matthias Weiß, Falk Dettinger, Michael Weyrich

Connected and software-defined vehicles promise to offer a broad range of services and advanced functions to customers, aiming to increase passenger comfort and support autonomous driving capabilities. Due to the high reliability and availability requirements of connected vehicles, it is crucial to resolve any occurring failures quickly. To achieve this however, a complex cloud/edge architecture with a mesh of dependencies must be navigated to diagnose the responsible root cause. As such, manual analyses become unfeasible since they would significantly delay the troubleshooting. To address this challenge, this paper presents SDVDiag, an extensible platform for the automated diagnosis of connected vehicle functions. The platform enables the creation of pipelines that cover all steps from initial data collection to the tracing of potential root causes. In addition, SDVDiag supports self-adaptive behavior by the ability to exchange modules at runtime. Dependencies between functions are detected and continuously updated, resulting in a dynamic graph view of the system. In addition, vital system metrics are monitored for anomalies. Whenever an incident is investigated, a snapshot of the graph is taken and augmented by relevant anomalies. Finally, the analysis is performed by traversing the graph and creating a ranking of the most likely causes. To evaluate the platform, it is deployed inside an 5G test fleet environment for connected vehicle functions. The results show that injected faults can be detected reliably. As such, the platform offers the potential to gain new insights and reduce downtime by identifying problems and their causes at an early stage.

Hannes Vietz, Tristan Rauch, Andreas Löcklin et al.

Developing consistently well performing visual recognition applications based on convolutional neural networks, e.g. for autonomous driving, is very challenging. One of the obstacles during the development is the opaqueness of their cognitive behaviour. A considerable amount of literature has been published which describes irrational behaviour of trained CNNs showcasing gaps in their cognition. In this paper, a methodology is presented that creates worstcase images using image augmentation techniques. If the CNN's cognitive performance on such images is weak while the augmentation techniques are supposedly harmless, a potential gap in the cognition has been found. The presented worst-case image generator is using adversarial search approaches to efficiently identify the most challenging image. This is evaluated with the well-known AlexNet CNN using images depicting a typical driving scenario.

Benjamin Maschler, Sophia Tatiyosyan, Michael Weyrich

In recent years, the use of lithium-ion batteries has greatly expanded into products from many industrial sectors, e.g. cars, power tools or medical devices. An early prediction and robust understanding of battery faults could therefore greatly increase product quality in those fields. While current approaches for data-driven fault prediction provide good results on the exact processes they were trained on, they often lack the ability to flexibly adapt to changes, e.g. in operational or environmental parameters. Continual learning promises such flexibility, allowing for an automatic adaption of previously learnt knowledge to new tasks. Therefore, this article discusses different continual learning approaches from the group of regularization strategies, which are implemented, evaluated and compared based on a real battery wear dataset. Online elastic weight consolidation delivers the best results, but, as with all examined approaches, its performance appears to be strongly dependent on task characteristics and task sequence.

Timo Müller, Benjamin Lindemann, Tobias Jung et al.

Shorter product life cycles and increasing individualization of production leads to an increased reconfiguration demand in the domain of industrial automation systems, which will be dominated by cyber-physical production systems in the future. In constantly changing systems, however, not all configuration alternatives of the almost infinite state space are fully understood. Thus, certain configurations can lead to process instability, a reduction in quality or machine failures. Therefore, this paper presents an approach that enhances an intelligent Digital Twin with a self-organized reconfiguration management based on adaptive process models in order to find optimized configurations more comprehensively.

Benjamin Maschler, Tim Knodel, Michael Weyrich

Deep learning promises performant anomaly detection on time-variant datasets, but greatly suffers from low availability of suitable training datasets and frequently changing tasks. Deep transfer learning offers mitigation by letting algorithms built upon previous knowledge from different tasks or locations. In this article, a modular deep learning algorithm for anomaly detection on time series datasets is presented that allows for an easy integration of such transfer learning capabilities. It is thoroughly tested on a dataset from a discrete manufacturing process in order to prove its fundamental adequacy towards deep industrial transfer learning - the transfer of knowledge in industrial applications' special environment.

Benjamin Maschler, Timo Müller, Andreas Löcklin et al.

Reconfiguration demand is increasing due to frequent requirement changes for manufacturing systems. Recent approaches aim at investigating feasible configuration alternatives from which they select the optimal one. This relies on processes whose behavior is not reliant on e.g. the production sequence. However, when machine learning is used, components' behavior depends on the process' specifics, requiring additional concepts to successfully conduct reconfiguration management. Therefore, we propose the enhancement of the comprehensive reconfiguration management with transfer learning. This provides the ability to assess the machine learning dependent behavior of the different CPPS configurations with reduced effort and further assists the recommissioning of the chosen one. A real cyber-physical production system from the discrete manufacturing domain is utilized to demonstrate the aforementioned proposal.

Benjamin Lindemann, Benjamin Maschler, Nada Sahlab et al.

Anomalies represent deviations from the intended system operation and can lead to decreased efficiency as well as partial or complete system failure. As the causes of anomalies are often unknown due to complex system dynamics, efficient anomaly detection is necessary. Conventional detection approaches rely on statistical and time-invariant methods that fail to address the complex and dynamic nature of anomalies. With advances in artificial intelligence and increasing importance for anomaly detection and prevention in various domains, artificial neural network approaches enable the detection of more complex anomaly types while considering temporal and contextual characteristics. In this article, a survey on state-of-the-art anomaly detection using deep neural and especially long short-term memory networks is conducted. The investigated approaches are evaluated based on the application scenario, data and anomaly types as well as further metrics. To highlight the potential of upcoming anomaly detection techniques, graph-based and transfer learning approaches are also included in the survey, enabling the analysis of heterogeneous data as well as compensating for its shortage and improving the handling of dynamic processes.

Benjamin Maschler, Thi Thu Huong Pham, Michael Weyrich

The early and robust detection of anomalies occurring in discrete manufacturing processes allows operators to prevent harm, e.g. defects in production machinery or products. While current approaches for data-driven anomaly detection provide good results on the exact processes they were trained on, they often lack the ability to flexibly adapt to changes, e.g. in products. Continual learning promises such flexibility, allowing for an automatic adaption of previously learnt knowledge to new tasks. Therefore, this article discusses different continual learning approaches from the group of regularization strategies, which are implemented, evaluated and compared based on a real industrial metal forming dataset.

Benjamin Maschler, Michael Weyrich

In this article, the concepts of transfer and continual learning are introduced. The ensuing review reveals promising approaches for industrial deep transfer learning, utilizing methods of both classes of algorithms. In the field of computer vision, it is already state-of-the-art. In others, e.g. fault prediction, it is barely starting. However, over all fields, the abstract differentiation between continual and transfer learning is not benefitting their practical use. In contrast, both should be brought together to create robust learning algorithms fulfilling the industrial automation sector's requirements. To better describe these requirements, base use cases of industrial transfer learning are introduced.

Benjamin Maschler, Dominik Braun, Nasser Jazdi et al.

Digital Twins have been described as beneficial in many areas, such as virtual commissioning, fault prediction or reconfiguration planning. Equipping Digital Twins with artificial intelligence functionalities can greatly expand those beneficial applications or open up altogether new areas of application, among them cross-phase industrial transfer learning. In the context of machine learning, transfer learning represents a set of approaches that enhance learning new tasks based upon previously acquired knowledge. Here, knowledge is transferred from one lifecycle phase to another in order to reduce the amount of data or time needed to train a machine learning algorithm. Looking at common challenges in developing and deploying industrial machinery with deep learning functionalities, embracing this concept would offer several advantages: Using an intelligent Digital Twin, learning algorithms can be designed, configured and tested in the design phase before the physical system exists and real data can be collected. Once real data becomes available, the algorithms must merely be fine-tuned, significantly speeding up commissioning and reducing the probability of costly modifications. Furthermore, using the Digital Twin's simulation capabilities virtually injecting rare faults in order to train an algorithm's response or using reinforcement learning, e.g. to teach a robot, become practically feasible. This article presents several cross-phase industrial transfer learning use cases utilizing intelligent Digital Twins. A real cyber physical production system consisting of an automated welding machine and an automated guided vehicle equipped with a robot arm is used to illustrate the respective benefits.