Oliver Niggemann

Tim Rensmeyer, Benjamin Craig, Denis Kramer et al.

Even though Bayesian neural networks offer a promising framework for modeling uncertainty, active learning and incorporating prior physical knowledge, few applications of them can be found in the context of interatomic force modeling. One of the main challenges in their application to learning interatomic forces is the lack of suitable Monte Carlo Markov chain sampling algorithms for the posterior density, as the commonly used algorithms do not converge in a practical amount of time for many of the state-of-the-art architectures. As a response to this challenge, we introduce a new Monte Carlo Markov chain sampling algorithm in this paper which can circumvent the problems of the existing sampling methods. In addition, we introduce a new stochastic neural network model based on the NequIP architecture and demonstrate that, when combined with our novel sampling algorithm, we obtain predictions with state-of-the-art accuracy as well as a good measure of uncertainty.

Alexander Windmann, Henrik Steude, Oliver Niggemann

Deep learning (DL) models have seen increased attention for time series forecasting, yet the application on cyber-physical systems (CPS) is hindered by the lacking robustness of these methods. Thus, this study evaluates the robustness and generalization performance of DL architectures on multivariate time series data from CPS. Our investigation focuses on the models' ability to handle a range of perturbations, such as sensor faults and noise, and assesses their impact on overall performance. Furthermore, we test the generalization and transfer learning capabilities of these models by exposing them to out-of-distribution (OOD) samples. These include deviations from standard system operations, while the core dynamics of the underlying physical system are preserved. Additionally, we test how well the models respond to several data augmentation techniques, including added noise and time warping. Our experimental framework utilizes a simulated three-tank system, proposed as a novel benchmark for evaluating the robustness and generalization performance of DL algorithms in CPS data contexts. The findings reveal that certain DL model architectures and training techniques exhibit superior effectiveness in handling OOD samples and various perturbations. These insights have significant implications for the development of DL models that deliver reliable and robust performance in real-world CPS applications.

Alexander Windmann, Philipp Wittenberg, Gianluca Manca et al.

Cyber-physical system (CPS) forecasting models depend on sensor streams with noisy, biased, missing, or temporally misaligned readings, yet standard forecasting evaluation often selects models by nominal error without showing whether they remain robust under such faults. We introduce SensorFault-Bench, a shared CPS-grounded sensor-fault stress-test protocol for evaluating forecasting architectures and robustness-improvement methods, and an operational taxonomy organizing the method comparison. Across four real-world datasets and eight scored scenarios governed by a standardized severity model, it reports worst-scenario degradation, clean mean squared error (MSE), and worst-scenario fault-time MSE, separating relative robustness from absolute error. A disjoint fault-transfer split lets explicit fault-training methods train on adjacent fault families while evaluation uses separate benchmark scenarios. Empirically, forecasting architectures favored by clean MSE can degrade sharply under faults, and clean-MSE rankings can disagree with worst-scenario fault-time error rankings. Chronos-2, the evaluated zero-shot foundation-model representative, matches or trails the last-value naive forecaster in clean MSE on the two single-target datasets and has the largest worst-scenario degradation on ETTh1 and Traffic, where all channels are forecast targets. For the evaluated robustness-improvement method set, paired deltas show selective degradation reductions: projected gradient descent adversarial training and randomized training lead where value faults dominate observed degradation, while fault augmentation leads where availability faults dominate. SensorFault-Bench provides open-source code, documented data access, and reproduction and extension guides, so new datasets, architectures, and robustness-improvement methods can be evaluated under the same CPS sensor-fault robustness protocol.

Jan-Philipp Roche, Oliver Niggemann, Jens Friebe

Existing black box modeling approaches in machine learning suffer from a fixed input and output feature combination. In this paper, a new approach to reconstruct missing variables in a set of time series is presented. An autoencoder is trained as usual with every feature on both sides and the neural network parameters are fixed after this training. Then, the searched variables are defined as missing variables at the autoencoder input and optimized via automatic differentiation. This optimization is performed with respect to the available features loss calculation. With this method, different input and output feature combinations of the trained model can be realized by defining the searched variables as missing variables and reconstructing them. The combination can be changed without training the autoencoder again. The approach is evaluated on the base of a strongly nonlinear electrical component. It is working well for one of four variables missing and generally even for multiple missing variables.

Jan Lukas Augustin, Oliver Niggemann

Traditional model-based diagnosis relies on constructing explicit system models, a process that can be laborious and expertise-demanding. In this paper, we propose a novel framework that combines concepts of model-based diagnosis with deep graph structure learning. This data-driven approach leverages data to learn the system's underlying structure and provide dynamic observations, represented by two distinct graph adjacency matrices. Our work facilitates a seamless integration of graph structure learning with model-based diagnosis by making three main contributions: (i) redefining the constructs of system representation, observations, and faults (ii) introducing two distinct versions of a self-supervised graph structure learning model architecture and (iii) demonstrating the potential of our data-driven diagnostic method through experiments on a system of coupled oscillators.

Philipp Rosenthal, Niels Demke, Frank Mantwill et al.

In product design, a decomposition of the overall product function into a set of smaller, interacting functions is usually considered a crucial first step for any computer-supported design tool. Here, we propose a new approach for the decomposition of functions especially suited for later solutions based on Artificial Intelligence. The presented approach defines the decomposition problem in terms of a planning problem--a well established field in Artificial Intelligence. For the planning problem, logic-based solvers can be used to find solutions that compute a useful function structure for the design process. Well-known function libraries from engineering are used as atomic planning steps. The algorithms are evaluated using two different application examples to ensure the transferability of a general function decomposition.

Maria Krantz, Alexander Windmann, Rene Heesch et al.

The increasing complexity of Cyber-Physical Systems (CPS) makes industrial automation challenging. Large amounts of data recorded by sensors need to be processed to adequately perform tasks such as diagnosis in case of fault. A promising approach to deal with this complexity is the concept of causality. However, most research on causality has focused on inferring causal relations between parts of an unknown system. Engineering uses causality in a fundamentally different way: complex systems are constructed by combining components with known, controllable behavior. As CPS are constructed by the second approach, most data-based causality models are not suited for industrial automation. To bridge this gap, a Uniform Causality Model for various application areas of industrial automation is proposed, which will allow better communication and better data usage across disciplines. The resulting model describes the behavior of CPS mathematically and, as the model is evaluated on the unique requirements of the application areas, it is shown that the Uniform Causality Model can work as a basis for the application of new approaches in industrial automation that focus on machine learning.

Lukas Moddemann, Henrik Sebastian Steude, Alexander Diedrich et al.

Consistency-based diagnosis is an established approach to diagnose technical applications, but suffers from significant modeling efforts, especially for dynamic multi-modal time series. Machine learning seems to be an obvious solution, which becomes less obvious when looking at details: Which notion of consistency can be used? If logical calculi are still to be used, how can dynamic time series be transferred into the discrete world? This paper presents the methodology Discret2Di for automated learning of logical expressions for consistency-based diagnosis. While these logical calculi have advantages by providing a clear notion of consistency, they have the key problem of relying on a discretization of the dynamic system. The solution presented combines machine learning from both the time series and the symbolic domain to automate the learning of logical rules for consistency-based diagnosis.

Felix Leonhard Janzen, Lukas Moddemann, Alexander Diedrich et al.

In many cyber-physical systems, especially in critical applications such as aeroplanes, data to train anomaly detection and diagnosis algorithms is lacking due to data protection issues and partial observability. To combat this inherent lack of data, we introduce a high-fidelity, physics-informed co-simulation of a common aircraft main-fuel-pump system modelled in \textsc{MATLAB/Simulink Simscape Fluids}. We also describe its generated time-series data with health and fault mode annotations. To show feasibility of our benchmark, we apply an unsupervised Recurrent Variational Autoencoder (RNN-VAE) for anomaly detection and a SOM-VAE for operating mode discretization, trained to separate healthy and faulty conditions.

Henrik S. Steude, Lukas Moddemann, Alexander Diedrich et al.

In Cyber-Physical Systems (CPS) research, anomaly detection (detecting abnormal behavior) and diagnosis (identifying the underlying root cause) are often treated as distinct, isolated tasks. However, diagnosis algorithms require symptoms, i.e. temporally and spatially isolated anomalies, as input. Thus, anomaly detection and diagnosis must be developed together to provide a holistic solution for diagnosis in CPS. We therefore propose a method for utilizing deep learning-based anomaly detection to generate inputs for Consistency-Based Diagnosis (CBD). We evaluate our approach on a simulated and a real-world CPS dataset, where our model demonstrates strong performance relative to other state-of-the-art models.

Lisa Lüddecke, Michael Hohmann, Sebastian Eilermann et al.

Reliable prediction of hydraulic performance is challenging for Piano Key Weir (PKW) design because discharge capacity depends on three-dimensional geometry and operating conditions. Surrogate models can accelerate hydraulic-structure design, but progress is limited by scarce large, well-documented datasets that jointly capture geometric variation, operating conditions, and functional performance. This study presents WeirNet, a large 3D CFD benchmark dataset for geometric surrogate modeling of PKWs. WeirNet contains 3,794 parametric, feasibility-constrained rectangular and trapezoidal PKW geometries, each scheduled at 19 discharge conditions using a consistent free-surface OpenFOAM workflow, resulting in 71,387 completed simulations that form the benchmark and with complete discharge coefficient labels. The dataset is released as multiple modalities compact parametric descriptors, watertight surface meshes and high-resolution point clouds together with standardized tasks and in-distribution and out-of-distribution splits. Representative surrogate families are benchmarked for discharge coefficient prediction. Tree-based regressors on parametric descriptors achieve the best overall accuracy, while point- and mesh-based models remain competitive and offer parameterization-agnostic inference. All surrogates evaluate in milliseconds per sample, providing orders-of-magnitude speedups over CFD runtimes. Out-of-distribution results identify geometry shift as the dominant failure mode compared to unseen discharge values, and data-efficiency experiments show diminishing returns beyond roughly 60% of the training data. By publicly releasing the dataset together with simulation setups and evaluation pipelines, WeirNet establishes a reproducible framework for data-driven hydraulic modeling and enables faster exploration of PKW designs during the early stages of hydraulic planning.

Christoph Petroll, Sebastian Eilermann, Philipp Hoefer et al.

One of the most promising developments in computer vision in recent years is the use of generative neural networks for functionality condition-based 3D design reconstruction and generation. Here, neural networks learn dependencies between functionalities and a geometry in a very effective way. For a neural network the functionalities are translated in conditions to a certain geometry. But the more conditions the design generation needs to reflect, the more difficult it is to learn clear dependencies. This leads to a multi criteria design problem due various conditions, which are not considered in the neural network structure so far. In this paper, we address this multi-criteria challenge for a 3D design use case related to an unmanned aerial vehicle (UAV) motor mount. We generate 10,000 abstract 3D designs and subject them all to simulations for three physical disciplines: mechanics, thermodynamics, and aerodynamics. Then, we train a Conditional Variational Autoencoder (CVAE) using the geometry and corresponding multicriteria functional constraints as input. We use our trained CVAE as well as the Marching cubes algorithm to generate meshes for simulation based evaluation. The results are then evaluated with the generated UAV designs. Subsequently, we demonstrate the ability to generate optimized designs under self-defined functionality conditions using the trained neural network.

Willi Grossmann, Sebastian Eilermann, Tim Rensmeyer et al.

Traditional design cycles for new materials and assemblies have two fundamental drawbacks. The underlying physical relationships are often too complex to be precisely calculated and described. Aside from that, many unknown uncertainties, such as exact manufacturing parameters or materials composition, dominate the real assembly behavior. Machine learning (ML) methods overcome these fundamental limitations through data-driven learning. In addition, modern approaches can specifically increase system knowledge. Representation Learning allows the physical, and if necessary, even symbolic interpretation of the learned solution. In this way, the most complex physical relationships can be considered and quickly described. Furthermore, generative ML approaches can synthesize possible morphologies of the materials based on defined conditions to visualize the effects of uncertainties. This modern approach accelerates the design process for new materials and enables the prediction and interpretation of realistic materials behavior.

Tim Rensmeyer, Denis Kramer, Oliver Niggemann

Due to the computational complexity of evaluating interatomic forces from first principles, the creation of interatomic machine learning force fields has become a highly active field of research. However, the generation of training datasets of sufficient size and sample diversity itself comes with a computational burden that can make this approach impractical for modeling rare events or systems with a large configuration space. Fine-tuning foundation models that have been pre-trained on large-scale material or molecular databases offers a promising opportunity to reduce the amount of training data necessary to reach a desired level of accuracy. However, even if this approach requires less training data overall, creating a suitable training dataset can still be a very challenging problem, especially for systems with rare events and for end-users who don't have an extensive background in machine learning. In on-the-fly learning, the creation of a training dataset can be largely automated by using model uncertainty during the simulation to decide if the model is accurate enough or if a structure should be recalculated with classical methods and used to update the model. A key challenge for applying this form of active learning to the fine-tuning of foundation models is how to assess the uncertainty of those models during the fine-tuning process, even though most foundation models lack any form of uncertainty quantification. In this paper, we overcome this challenge by introducing a fine-tuning approach based on Bayesian neural network methods and a subsequent on-the-fly workflow that automatically fine-tunes the model while maintaining a pre-specified accuracy and can detect rare events such as transition states and sample them at an increased rate relative to their occurrence.

Rene Heesch, Sebastian Eilermann, Alexander Windmann et al.

Large Language Models (LLMs) are transformative not only for daily activities but also for engineering tasks. However, current evaluations of LLMs in engineering exhibit two critical shortcomings: (i) the reliance on simplified use cases, often adapted from examination materials where correctness is easily verifiable, and (ii) the use of ad hoc scenarios that insufficiently capture critical engineering competencies. Consequently, the assessment of LLMs on complex, real-world engineering problems remains largely unexplored. This paper addresses this gap by introducing a curated database comprising over 100 questions derived from authentic, production-oriented engineering scenarios, systematically designed to cover core competencies such as product design, prognosis, and diagnosis. Using this dataset, we evaluate four state-of-the-art LLMs, including both cloud-based and locally hosted instances, to systematically investigate their performance on complex engineering tasks. Our results show that LLMs demonstrate strengths in basic temporal and structural reasoning but struggle significantly with abstract reasoning, formal modeling, and context-sensitive engineering logic.

Tim Rensmeyer, Oliver Niggemann

Achieving robust uncertainty quantification for deep neural networks represents an important requirement in many real-world applications of deep learning such as medical imaging where it is necessary to assess the reliability of a neural network's prediction. Bayesian neural networks are a promising approach for modeling uncertainties in deep neural networks. Unfortunately, generating samples from the posterior distribution of neural networks is a major challenge. One significant advance in that direction would be the incorporation of adaptive step sizes, similar to modern neural network optimizers, into Monte Carlo Markov chain sampling algorithms without significantly increasing computational demand. Over the past years, several papers have introduced sampling algorithms with claims that they achieve this property. However, do they indeed converge to the correct distribution? In this paper, we demonstrate that these methods can have a substantial bias in the distribution they sample, even in the limit of vanishing step sizes and at full batch size.

Alexander Windmann, Benedikt Stratmann, Mariya Lyashenko et al.

Industrial AI practitioners face vague robustness requirements in emerging regulations and standards but lack concrete, implementation ready protocols. This paper introduces the Industrial AI Robustness Card (IARC), a lightweight, task agnostic protocol for documenting and evaluating the robustness of AI models on industrial time series. The IARC specifies required fields and an empirical measurement and reporting protocol that combines drift monitoring, uncertainty quantification, and stress tests, and it maps these to relevant EU AI Act obligations. A soft sensor case study on a biopharmaceutical fermentation process illustrates how the IARC supports reproducible robustness evidence and continuous monitoring.

Hamied Nabizada, Lasse Beers, Alain Chahine et al.

Engineering models created in Model-Based Systems Engineering (MBSE) environments contain detailed information about system structure and behavior. However, they typically lack symbolic planning semantics such as preconditions, effects, and constraints related to resource availability and timing. This limits their ability to evaluate whether a given system variant can fulfill specific tasks and how efficiently it performs compared to alternatives. To address this gap, this paper presents a model-driven method that enables the specification and automated generation of symbolic planning artifacts within SysML-based engineering models. A dedicated SysML profile introduces reusable stereotypes for core planning constructs. These are integrated into existing model structures and processed by an algorithm that generates a valid domain file and a corresponding problem file in Planning Domain Definition Language (PDDL). In contrast to previous approaches that rely on manual transformations or external capability models, the method supports native integration and maintains consistency between engineering and planning artifacts. The applicability of the method is demonstrated through a case study from aircraft assembly. The example illustrates how existing engineering models are enriched with planning semantics and how the proposed workflow is applied to generate consistent planning artifacts from these models. The generated planning artifacts enable the validation of system variants through AI planning.

Sebastian Eilermann, René Heesch, Oliver Niggemann

Fast and accurate 3D shape generation from point clouds is essential for applications in robotics, AR/VR, and digital content creation. We introduce ConTiCoM-3D, a continuous-time consistency model that synthesizes 3D shapes directly in point space, without discretized diffusion steps, pre-trained teacher models, or latent-space encodings. The method integrates a TrigFlow-inspired continuous noise schedule with a Chamfer Distance-based geometric loss, enabling stable training on high-dimensional point sets while avoiding expensive Jacobian-vector products. This design supports efficient one- to two-step inference with high geometric fidelity. In contrast to previous approaches that rely on iterative denoising or latent decoders, ConTiCoM-3D employs a time-conditioned neural network operating entirely in continuous time, thereby achieving fast generation. Experiments on the ShapeNet benchmark show that ConTiCoM-3D matches or outperforms state-of-the-art diffusion and latent consistency models in both quality and efficiency, establishing it as a practical framework for scalable 3D shape generation.

Joshua Schraven, Alexander Windmann, Oliver Niggemann

Benchmark datasets for network intrusion detection commonly rely on synthetically generated traffic, which fails to reflect the statistical variability and temporal drift encountered in operational environments. This paper introduces MAWIFlow, a flow-based benchmark derived from the MAWILAB v1.1 dataset, designed to enable realistic and reproducible evaluation of anomaly detection methods. A reproducible preprocessing pipeline is presented that transforms raw packet captures into flow representations conforming to the CICFlowMeter format, while preserving MAWILab's original anomaly labels. The resulting datasets comprise temporally distinct samples from January 2011, 2016, and 2021, drawn from trans-Pacific backbone traffic. To establish reference baselines, traditional machine learning methods, including Decision Trees, Random Forests, XGBoost, and Logistic Regression, are compared to a deep learning model based on a CNN-BiLSTM architecture. Empirical results demonstrate that tree-based classifiers perform well on temporally static data but experience significant performance degradation over time. In contrast, the CNN-BiLSTM model maintains better performance, thus showing improved generalization. These findings underscore the limitations of synthetic benchmarks and static models, and motivate the adoption of realistic datasets with explicit temporal structure. All datasets, pipeline code, and model implementations are made publicly available to foster transparency and reproducibility.

Henrik Sebastian Steude, Alexander Diedrich, Ingo Pill et al.

Diagnostic processes for complex cyber-physical systems often require extensive prior knowledge in the form of detailed system models or comprehensive training data. However, obtaining such information poses a significant challenge. To address this issue, we present a new diagnostic approach that operates with minimal prior knowledge, requiring only a basic understanding of subsystem relationships and data from nominal operations. Our method combines a neural network-based symptom generator, which employs subsystem-level anomaly detection, with a new graph diagnosis algorithm that leverages minimal causal relationship information between subsystems-information that is typically available in practice. Our experiments with fully controllable simulated datasets show that our method includes the true causal component in its diagnosis set for 82 p.c. of all cases while effectively reducing the search space in 73 p.c. of the scenarios. Additional tests on the real-world Secure Water Treatment dataset showcase the approach's potential for practical scenarios. Our results thus highlight our approach's potential for practical applications with large and complex cyber-physical systems where limited prior knowledge is available.

Alexander Windmann, Henrik Steude, Daniel Boschmann et al.

Cyber-Physical Systems (CPS) in domains such as manufacturing and energy distribution generate complex time series data crucial for Prognostics and Health Management (PHM). While Deep Learning (DL) methods have demonstrated strong forecasting capabilities, their adoption in industrial CPS remains limited due insufficient robustness. Existing robustness evaluations primarily focus on formal verification or adversarial perturbations, inadequately representing the complexities encountered in real-world CPS scenarios. To address this, we introduce a practical robustness definition grounded in distributional robustness, explicitly tailored to industrial CPS, and propose a systematic framework for robustness evaluation. Our framework simulates realistic disturbances, such as sensor drift, noise and irregular sampling, enabling thorough robustness analyses of forecasting models on real-world CPS datasets. The robustness definition provides a standardized score to quantify and compare model performance across diverse datasets, assisting in informed model selection and architecture design. Through extensive empirical studies evaluating prominent DL architectures (including recurrent, convolutional, attention-based, modular, and structured state-space models) we demonstrate the applicability and effectiveness of our approach. We publicly release our robustness benchmark to encourage further research and reproducibility.

Bernd Zimmering, Cecília Coelho, Vaibhav Gupta et al.

Modelling forced dynamical systems - where an external input drives the system state - is critical across diverse domains such as engineering, finance, and the natural sciences. In this work, we propose Laplace-Net, a decoupled, solver-free neural framework for learning forced and delay-aware systems. It leverages a Laplace transform-based approach to decompose internal dynamics, external inputs, and initial values into established theoretical concepts, enhancing interpretability. Laplace-Net promotes transferability since the system can be rapidly re-trained or fine-tuned for new forcing signals, providing flexibility in applications ranging from controller adaptation to long-horizon forecasting. Experimental results on eight benchmark datasets - including linear, non-linear, and delayed systems - demonstrate the method's improved accuracy and robustness compared to state-of-the-art approaches, particularly in handling complex and previously unseen inputs.

Maria Krantz, Oliver Niggemann

Rotary Indexing Machines (RIMs) are widely used in manufacturing due to their ability to perform multiple production steps on a single product without manual repositioning, reducing production time and improving accuracy and consistency. Despite their advantages, little research has been done on diagnosing faults in RIMs, especially from the perspective of the actual production steps carried out on these machines. Long downtimes due to failures are problematic, especially for smaller companies employing these machines. To address this gap, we propose a diagnosis algorithm based on the product perspective, which focuses on the product being processed by RIMs. The algorithm traces the steps that a product takes through the machine and is able to diagnose possible causes in case of failure. We also analyze the properties of RIMs and how these influence the diagnosis of faults in these machines. Our contributions are three-fold. Firstly, we provide an analysis of the properties of RIMs and how they influence the diagnosis of faults in these machines. Secondly, we suggest a diagnosis algorithm based on the product perspective capable of diagnosing faults in such a machine. Finally, we test this algorithm on a model of a rotary indexing machine, demonstrating its effectiveness in identifying faults and their root causes.

Daniel C. Hinck, Jonas J. Schöttler, Maria Krantz et al.

The protection of non-combatants in times of (fully) autonomous warfare raises the question of the timeliness of the international protective emblem. Incidents in the recent past indicate that it is becoming necessary to transfer the protective emblem to other dimensions of transmission and representation. (Fully) Autonomous weapon systems are often launched from a great distance to the aiming point and there may be no possibility for the operators to notice protective emblems at the point of impact. In this case, the weapon system would have to detect such protective emblems and, if necessary, disintegrate autonomously or request an abort via human-in-the-loop. In our paper, we suggest ways in which a cross-frequency protective emblem can be designed. On the one hand, the technical deployment, e.g. in the form of RADAR beacons, is considered, as well as the interpretation by methods of machine learning. With regard to the technical deployment, possibilities are considered to address different sensors and to send signals out as resiliently as possible. When considering different signals, approaches are considered as to how software can recognise the protective emblems under the influence of various boundary conditions and react to them accordingly. In particular, a distinction is made here between the recognition of actively emitted signals and passive protective signals, e.g. the recognition of wounded or surrendering persons via drone-based electro-optical and thermal cameras. Finally, methods of distribution are considered, including encryption and authentication of the received signal, and ethical aspects of possible misuse are examined.

Philipp Rosenthal, Oliver Niggemann

Artificial Intelligence (AI) has significant potential for product design: AI can check technical and non-technical constraints on products, it can support a quick design of new product variants and new AI methods may also support creativity. But currently product design and AI are separate communities fostering different terms and theories. This makes a mapping of AI approaches to product design needs difficult and prevents new solutions. As a solution, this paper first clarifies important terms and concepts for the interdisciplinary domain of AI methods in product design. A key contribution of this paper is a new classification of design problems using the four characteristics decomposability, inter-dependencies, innovation and creativity. Definitions of these concepts are given where they are lacking. Early mappings of these concepts to AI solutions are sketched and verified using design examples. The importance of creativity in product design and a corresponding gap in AI is pointed out for future research.

Aljosha Köcher, Rene Heesch, Niklas Widulle et al.

Manufacturing companies face challenges when it comes to quickly adapting their production control to fluctuating demands or changing requirements. Control approaches that encapsulate production functions as services have shown to be promising in order to increase the flexibility of Cyber-Physical Production Systems. But an existing challenge of such approaches is finding a production plan based on provided functionalities for a demanded product, especially when there is no direct (i.e., syntactic) match between demanded and provided functions. While there is a variety of approaches to production planning, flexible production poses specific requirements that are not covered by existing research. In this contribution, we first capture these requirements for flexible production environments. Afterwards, an overview of current Artificial Intelligence approaches that can be utilized in order to overcome the aforementioned challenges is given. For this purpose, we focus on planning algorithms, but also consider models of production systems that can act as inputs to these algorithms. Approaches from both symbolic AI planning as well as approaches based on Machine Learning are discussed and eventually compared against the requirements. Based on this comparison, a research agenda is derived.

Henrik S. Steude, Alexander Windmann, Oliver Niggemann

Machine Learning (ML) has achieved great successes in recent decades, both in research and in practice. In Cyber-Physical Systems (CPS), ML can for example be used to optimize systems, to detect anomalies or to identify root causes of system failures. However, existing algorithms suffer from two major drawbacks: (i) They are hard to interpret by human experts. (ii) Transferring results from one systems to another (similar) system is often a challenge. Concept learning, or Representation Learning (RepL), is a solution to both of these drawbacks; mimicking the human solution approach to explain-ability and transfer-ability: By learning general concepts such as physical quantities or system states, the model becomes interpretable by humans. Furthermore concepts on this abstract level can normally be applied to a wide range of different systems. Modern ML methods are already widely used in CPS, but concept learning and transfer learning are hardly used so far. In this paper, we provide an overview of the current state of research regarding methods for learning physical concepts in time series data, which is the primary form of sensor data of CPS. We also analyze the most important methods from the current state of the art using the example of a three-tank system. Based on these concrete implementations1, we discuss the advantages and disadvantages of the methods and show for which purpose and under which conditions they can be used.

Kaja Balzereit, Oliver Niggemann

Reconfiguration aims at recovering a system from a fault by automatically adapting the system configuration, such that the system goal can be reached again. Classical approaches typically use a set of pre-defined faults for which corresponding recovery actions are defined manually. This is not possible for modern hybrid systems which are characterized by frequent changes. Instead, AI-based approaches are needed which leverage on a model of the non-faulty system and which search for a set of reconfiguration operations which will establish a valid behavior again. This work presents a novel algorithm which solves three main challenges: (i) Only a model of the non-faulty system is needed, i.e. the faulty behavior does not need to be modeled. (ii) It discretizes and reduces the search space which originally is too large -- mainly due to the high number of continuous system variables and control signals. (iii) It uses a SAT solver for propositional logic for two purposes: First, it defines the binary concept of validity. Second, it implements the search itself -- sacrificing the optimal solution for a quick identification of an arbitrary solution. It is shown that the approach is able to reconfigure faults on simulated process engineering systems.

Benedikt Eiteneuer, Oliver Niggemann

Anomaly detection is the task of detecting data which differs from the normal behaviour of a system in a given context. In order to approach this problem, data-driven models can be learned to predict current or future observations. Oftentimes, anomalous behaviour depends on the internal dynamics of the system and looks normal in a static context. To address this problem, the model should also operate depending on state. Long Short-Term Memory (LSTM) neural networks have been shown to be particularly useful to learn time sequences with varying length of temporal dependencies and are therefore an interesting general purpose approach to learn the behaviour of arbitrarily complex Cyber-Physical Systems. In order to perform anomaly detection, we slightly modify the standard norm 2 error to incorporate an estimate of model uncertainty. We analyse the approach on artificial and real data.

Nemanja Hranisavljevic, Oliver Niggemann, Alexander Maier

Performing anomaly detection in hybrid systems is a challenging task since it requires analysis of timing behavior and mutual dependencies of both discrete and continuous signals. Typically, it requires modeling system behavior, which is often accomplished manually by human engineers. Using machine learning for creating a behavioral model from observations has advantages, such as lower development costs and fewer requirements for specific knowledge about the system. The paper presents DAD:DeepAnomalyDetection, a new approach for automatic model learning and anomaly detection in hybrid production systems. It combines deep learning and timed automata for creating behavioral model from observations. The ability of deep belief nets to extract binary features from real-valued inputs is used for transformation of continuous to discrete signals. These signals, together with the original discrete signals are than handled in an identical way. Anomaly detection is performed by the comparison of actual and predicted system behavior. The algorithm has been applied to few data sets including two from real systems and has shown promising results.

Benedikt Eiteneuer, Nemanja Hranisavljevic, Oliver Niggemann

Unsupervised anomaly detection (AD) is a major topic in the field of Cyber-Physical Production Systems (CPPSs). A closely related concern is dimensionality reduction (DR) which is: 1) often used as a preprocessing step in an AD solution, 2) a sort of AD, if a measure of observation conformity to the learned data manifold is provided. We argue that the two aspects can be complementary in a CPPS anomaly detection solution. In this work, we focus on the nonlinear autoencoder (AE) as a DR/AD approach. The contribution of this work is: 1) we examine the suitability of AE reconstruction error as an AD decision criterion in CPPS data. 2) we analyze its relation to a potential second-phase AD approach in the AE latent space 3) we evaluate the performance of the approach on three real-world datasets. Moreover, the approach outperforms state-of-the-art techniques, alongside a relatively simple and straightforward application.

Oliver Niggemann, Alexander Diedrich, Christian Kuehnert et al.

Services for Cyber-Physical Systems based on Artificial Intelligence and Machine Learning require a virtual representation of the physical. To reduce modeling efforts and to synchronize results, for each system, a common and unique virtual representation used by all services during the whole system life-cycle is needed, i.e. a DigitalTwin. In this paper such a DigitalTwin, namely the AI reference model AITwin, is defined. This reference model is verified by using a running example from process industry and by analyzing the work done in recent projects.