Search

Wireless communication between road users is essential for environmental perception, reasoning, and mission planning to enable fully autonomous vehicles, and thus improve road safety and transport efficiency. To enable collaborative driving, the concept of vehicle-to-Everything (V2X) has long been introduced to the industry. Within the last two decades, several communication standards have been developed based on IEEE 802.11p and cellular standards, namely Dedicated Short-Range Communication (DSRC), Intelligent Transportation System G5 (ITS-G5), and Cellular- and New Radio- Vehicle-to-Everything (C-V2X and NR-V2X). However, while there exists a high quantity of available publications concerning V2X and the analysis of the different standards, only few surveys exist that summarize these results. Furthermore, to our knowledge, no survey that provides an analysis about possible future trends and challenges for the global implementation of V2Xexists. Thus, this contribution provides a detailed survey on Vehicle-to-Everything communication standards, their performance, current and future applications, and associated challenges. Based on our research, we have identified several research gaps and provide a picture about the possible future of the Vehicle-to-Everything communication domain.

3D Gaussian Splats (3DGS) have proven a versatile rendering primitive, both for inverse rendering as well as real-time exploration of scenes. In these applications, coherence across camera frames and multiple views is crucial, be it for robust convergence of a scene reconstruction or for artifact-free fly-throughs. Recent work started mitigating artifacts that break multi-view coherence, including popping artifacts due to inconsistent transparency sorting and perspective-correct outlines of (2D) splats. At the same time, real-time requirements forced such implementations to accept compromises in how transparency of large assemblies of 3D Gaussians is resolved, in turn breaking coherence in other ways. In our work, we aim at achieving maximum coherence, by rendering fully perspective-correct 3D Gaussians while using a high-quality approximation of accurate blending, hybrid transparency, on a per-pixel level, in order to retain real-time frame rates. Our fast and perspectively accurate approach for evaluation of 3D Gaussians does not require matrix inversions, thereby ensuring numerical stability and eliminating the need for special handling of degenerate splats, and the hybrid transparency formulation for blending maintains similar quality as fully resolved per-pixel transparencies at a fraction of the rendering costs. We further show that each of these two components can be independently integrated into Gaussian splatting systems. In combination, they achieve up to 2$\times$ higher frame rates, 2$\times$ faster optimization, and equal or better image quality with fewer rendering artifacts compared to traditional 3DGS on common benchmarks., Comment: Project page: https://fhahlbohm.github.io/htgs/

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

Achieving photorealistic 3D view synthesis and relighting of human portraits is pivotal for advancing AR/VR applications. Existing methodologies in portrait relighting demonstrate substantial limitations in terms of generalization and 3D consistency, coupled with inaccuracies in physically realistic lighting and identity preservation. Furthermore, personalization from a single view is difficult to achieve and often requires multiview images during the testing phase or involves slow optimization processes. This paper introduces Lite2Relight, a novel technique that can predict 3D consistent head poses of portraits while performing physically plausible light editing at interactive speed. Our method uniquely extends the generative capabilities and efficient volumetric representation of EG3D, leveraging a lightstage dataset to implicitly disentangle face reflectance and perform relighting under target HDRI environment maps. By utilizing a pre-trained geometry-aware encoder and a feature alignment module, we map input images into a relightable 3D space, enhancing them with a strong face geometry and reflectance prior. Through extensive quantitative and qualitative evaluations, we show that our method outperforms the state-of-the-art methods in terms of efficacy, photorealism, and practical application. This includes producing 3D-consistent results of the full head, including hair, eyes, and expressions. Lite2Relight paves the way for large-scale adoption of photorealistic portrait editing in various domains, offering a robust, interactive solution to a previously constrained problem. Project page: https://vcai.mpi-inf.mpg.de/projects/Lite2Relight/, Comment: Accepted at SIGGRAPH '24: ACM SIGGRAPH 2024 Conference Papers

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

Software product lines (SPL) have emerged as a pivotal paradigm in software engineering, enabling the efficient development of variant-rich software systems. Consistently updating these systems, often through over-the-air updates, enables the continuous integration of new features and bug fixes, ensuring the system remains up to date throughout its entire lifecycle. However, evolving such complex systems is an error prone task, leading to a phenomenon known as erosion. This phenomenon significantly impacts the efficiency and longevity of software systems, presenting a formidable challenge for manufacturers of variant-rich software systems, such as in the automotive domain. While existing studies concentrate on the evolutionary planning of variant-rich software systems, there is a noticeable lack of research addressing the problem of erosion. In this paper, we conduct an in-depth exploration of the erosion phenomena within variant-rich software systems. We begin by highlighting the significance of controlling erosion in extensive variant-rich software systems. Subsequently, we address the current challenges regarding tackling erosion, including issues such as the lack of a consensus on understanding and defining erosion, as well as the early detection and elimination. Finally, we outline a first approach aimed at tackling erosion in variant-rich software systems.

The ongoing digitization of the industrial sector has reached a pivotal juncture with the emergence of Digital Twins, offering a digital representation of physical assets and processes. One key aspect of those digital representations are simulation models, enabling a deeper insight in the assets current state and its characteristics. This paper asserts that the next evolutionary step in this digitization journey involves the integration of intelligent linkages between diverse simulation models within the Digital Twin framework. Crucially, for the Digital Twin to be a scalable and cost-effective solution, there is a pressing need for automated adaption, (re-)configuration, and generation of simulation models. Recognizing the inherent challenges in achieving such automation, this paper analyses the utilization of knowledge graphs as a potentially very suitable technological solution. Knowledge graphs, acting as interconnected and interrelated databases, provide a means of seamlessly integrating different data sources, facilitating the efficient integration and automated adaption of data and (simulation) models in the Digital Twin. We conducted a comprehensive literature review to analyze the current landscape of knowledge graphs in the context of Digital Twins with focus on simulation models. By addressing the challenges associated with scalability and maintenance, this research contributes to the effective adaption of Digital Twins in the industrial sector, paving the way for enhanced efficiency, adaptability, and resilience in the face of evolving technological landscapes.

The concept of the Digital Twin, which in the context of this paper is the virtual representation of a production system or its components, can be used as a "digital playground" to master the increasing complexity of these assets. One of the central subcomponents of the Digital Twin are behavior models that can enable benefits over the entire machine life cycle. However, the creation, adaption and use of behavior models throughout the machine life cycle is very time-consuming, which is why approaches to improve the cost-benefit ratio are needed. Furthermore, there is a lack of specific use cases that illustrate the application and added benefit of behavior models over the machine life cycle, which is why the universal application of behavior models in industry is still lacking compared to research. This paper first presents the fundamentals, challenges and related work on Digital Twins and behavior models in the context of the machine life cycle. Then, concepts for low-effort creation and automatic adaption of Digital Twins are presented, with a focus on behavior models. Finally, the aforementioned gap between research and industry is addressed by demonstrating various realized use cases over the machine life cycle, in which the advantages as well as the application of behavior models in the different life cycle phases are shown.

This paper presents a novel design of a multi-agent system framework that applies large language models (LLMs) to automate the parametrization of simulation models in digital twins. This framework features specialized LLM agents tasked with observing, reasoning, decision-making, and summarizing, enabling them to dynamically interact with digital twin simulations to explore parametrization possibilities and determine feasible parameter settings to achieve an objective. The proposed approach enhances the usability of simulation model by infusing it with knowledge heuristics from LLM and enables autonomous search for feasible parametrization to solve a user task. Furthermore, the system has the potential to increase user-friendliness and reduce the cognitive load on human users by assisting in complex decision-making processes. The effectiveness and functionality of the system are demonstrated through a case study, and the visualized demos and codes are available at a GitHub Repository: https://github.com/YuchenXia/LLMDrivenSimulation, Comment: Submitted to IEEE-ETFA2024, under peer-review

Anomaly detection and localization in images is a growing field in computer vision. In this area, a seemingly understudied problem is anomaly clustering, i.e., identifying and grouping different types of anomalies in a fully unsupervised manner. In this work, we propose a novel method for clustering anomalies in largely stationary images (textures) in a blind setting. That is, the input consists of normal and anomalous images without distinction and without labels. What contributes to the difficulty of the task is that anomalous regions are often small and may present only subtle changes in appearance, which can be easily overshadowed by the genuine variance in the texture. Moreover, each anomaly type may have a complex appearance distribution. We introduce a novel scheme for solving this task using a combination of blind anomaly localization and contrastive learning. By identifying the anomalous regions with high fidelity, we can restrict our focus to those regions of interest; then, contrastive learning is employed to increase the separability of different anomaly types and reduce the intra-class variation. Our experiments show that the proposed solution yields significantly better results compared to prior work, setting a new state of the art. Project page: https://reality.tf.fau.de/pub/ardelean2024blind.html.

The emergence of connected vehicles is driven by increasing customer and regulatory demands. To meet these, more complex software applications, some of which require service-based cloud and edge backends, are developed. When new software is deployed however, the high complexity and interdependencies between components can lead to unforeseen side effects in other system parts. As such, it becomes more challenging to recognize whether deviations to the intended system behavior are occurring, ultimately resulting in higher monitoring efforts and slower responses to errors. To overcome this problem, a simulation of the cloud environment running in parallel to the system is proposed. This approach enables the live comparison between simulated and real cloud behavior. Therefore, a concept is developed mirroring the existing cloud system into a simulation. To collect the necessary data, an observability platform is presented, capturing telemetry and architecture information. Subsequently, a simulation environment is designed that converts the architecture into a simulation model and simulates its dynamic workload by utilizing captured communication data. The proposed concept is evaluated in a real-world application scenario for electric vehicle charging: Vehicles can apply for an unoccupied charging station at a cloud service backend, the latter which manages all incoming requests and performs the assignment. Benchmarks are conducted by comparing the collected telemetry data with the simulated results under different loads and injected faults. The results show that regular cloud behavior is mirrored well by the simulation and that misbehavior due to fault injection is well visible, indicating that simulations are a promising data source for anomaly detection in connected vehicle cloud environments during operation., Comment: 11 pages, 10 figures

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., Comment: Published in IEEE Access

Despite an emerging interest in MIMO radar, the utilization of its complementary strengths in combination with optical depth sensors has so far been limited to far-field applications, due to the challenges that arise from mutual sensor calibration in the near field. In fact, most related approaches in the autonomous industry propose target-based calibration methods using corner reflectors that have proven to be unsuitable for the near field. In contrast, we propose a novel, joint calibration approach for optical RGB-D sensors and MIMO radars that is designed to operate in the radar's near-field range, within decimeters from the sensors. Our pipeline consists of a bespoke calibration target, allowing for automatic target detection and localization, followed by the spatial calibration of the two sensor coordinate systems through target registration. We validate our approach using two different depth sensing technologies from the optical domain. The experiments show the efficiency and accuracy of our calibration for various target displacements, as well as its robustness of our localization in terms of signal ambiguities., Comment: 8 pages, 9 figures, accepted to IROS 2024

The Digital Twin is a well-known concept of industry 4.0 and is the cyber part of a cyber-physical production system providing several benefits such as virtual commissioning or predictive maintenance. The existing production systems are lacking a Digital Twin which has to be created manually in a time-consuming and error-prone process. Therefore, methods to create digital models of existing production systems and their relations between them were developed. This paper presents the implementation of the methodology for the creation of multi-disciplinary relations and a quantitative and qualitative evaluation of the benefits of the methodology., Comment: Preprint, 7 pages, 8 figures, Accepted at IEEE ETFA 2023

Power-to-X (PtX) products constitute a promising solution component in the defossilisation of hard-to-abate sectors. Where direct electrification is not possible, they can find application as chemical feedstock, easy to transport energy carriers and storage. In this contribution, a modular system architecture for a highly automated operation of an offshore off-grid PtX production platform consisting of five layers (asset, control, communication, digital, application) is presented. The assumed advantage of such an off-grid platform is the good scalability worldwide as well as easy exploitation of renewable potentials in any location. Using the example of a power-to-methanol process, a simplified prototype is outlined., Comment: 6 pages, 3 figures. Accepted at CIRP ICME 2023

In this work a novel radar simulation concept is introduced that allows to simulate realistic radar data for Range, Doppler, and for arbitrary antenna positions in an efficient way. Further, it makes it possible to automatically annotate the simulated radar signal by allowing to decompose it into different parts. This approach allows not only almost perfect annotations possible, but also allows the annotation of exotic effects, such as multi-path effects or to label signal parts originating from different parts of an object. This is possible by adapting the computation process of a Monte Carlo shooting and bouncing rays (SBR) simulator. By considering the hits of each simulated ray, various meta data can be stored such as hit position, mesh pointer, object IDs, and many more. This collected meta data can then be utilized to predict the change of path lengths introduced by object motion to obtain Doppler information or to apply specific ray filter rules in order obtain radar signals that only fulfil specific conditions, such as multiple bounces or containing specific object IDs. Using this approach, perfect and otherwise almost impossible annotations schemes can be realized., Comment: Accepted for IEEE RadarConf 2023

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, Comment: This is the pre-print draft manuscript. The peer-reviewed version will be published exclusively by IEEE after the conference, which is set to take place from September 12th to 15th, 2023. We've made several improvements to the final version of the paper based on valuable feedback and suggestions from other researchers

We propose a novel method for Zero-Shot Anomaly Localization on textures. The task refers to identifying abnormal regions in an otherwise homogeneous image. To obtain a high-fidelity localization, we leverage a bijective mapping derived from the 1-dimensional Wasserstein Distance. As opposed to using holistic distances between distributions, the proposed approach allows pinpointing the non-conformity of a pixel in a local context with increased precision. By aggregating the contribution of the pixel to the errors of all nearby patches we obtain a reliable anomaly score estimate. We validate our solution on several datasets and obtain more than a 40% reduction in error over the previous state of the art on the MVTec AD dataset in a zero-shot setting. Also see https://reality.tf.fau.de/pub/ardelean2024highfidelity.html.

The trend in industrial automation is towards networking, intelligence and autonomy. Digital Twins, which serve as virtual representations, are becoming increasingly important in this context. The Digital Twin of a modular production system contains many different models that are mostly created for specific applications and fulfil different requirements. Especially simulation models, which are created in the development phase, can be used during the operational phase for applications such as prognosis or operation-parallel simulation. Due to the high heterogeneity of the model landscape in the context of a modular production system, the plant operator is faced with the challenge of adapting the models in order to ensure an application-oriented realism in the event of changes to the asset and its environment or the addition of applications. Therefore, this paper proposes a concept for the continuous model adaption in the Digital Twin of a modular production system during the operational phase. The benefits are then demonstrated by an application scenario and an agent-based realisation., Comment: 6 pages, 5 figures

The creation of a Digital Twin for existing manufacturing systems, so-called brownfield systems, is a challenging task due to the needed expert knowledge about the structure of brownfield systems and the effort to realize the digital models. Several approaches and methods have already been proposed that at least partially digitalize the information about a brownfield manufacturing system. A Digital Twin requires linked information from multiple sources. This paper presents a graph-based approach to merge information from heterogeneous sources. Furthermore, the approach provides a way to automatically identify templates using graph structure analysis to facilitate further work with the resulting Digital Twin and its further enhancement., Comment: 4 pages, 3 figures. Accepted at IEEE ETFA 2022

The success of the reconfiguration of existing manufacturing systems, so called brownfield systems, heavily relies on the knowledge about the system. Reconfiguration can be planned, supported and simplified with the Digital Twin of the system providing this knowledge. However, digital models as the basis of a Digital Twin are usually missing for these plants. This article presents a data-driven approach to gain knowledge about a brownfield system to create the digital models of a Digital Twin and their relations. Finally, a proof of concept shows that process data and position data as data sources deliver the relations between the models of the Digital Twin., Comment: 7 Pages, 5 figures. Accepted at IEEE ETFA 2021

Systematic Literature Reviews aim at investigating current approaches to conclude a research gap or determine a futuristic approach. They represent a significant part of a research activity, from which new concepts stem. However, with the massive availability of publications at a rapid growing rate, especially digitally, it becomes challenging to efficiently screen and assess relevant publications. Another challenge is the continuous assessment of related work over a long period of time and the consequent need for a continuous update, which can be a time-consuming task. Knowledge graphs model entities in a connected manner and enable new insights using different reasoning and analysis methods. The objective of this work is to present an approach to partially automate the conduction of a Systematic Literature Review as well as classify and visualize the results as a knowledge graph. The designed software prototype was used for the conduction of a review on context-awareness in automation systems with considerably accurate results compared to a manual conduction., Comment: 9 pages, 7 figures, 2 tables

Automation systems are increasingly being used in dynamic and various operating conditions. With higher flexibility demands, they need to promptly respond to surrounding dynamic changes by adapting their operation. Context information collected during runtime can be useful to enhance the system's adaptability. Context-aware systems represent a design paradigm for modeling and applying context in various applications such as decision-making. In order to address context for automation systems, a state of the art assessment of existing approaches is necessary. Thus, the objective of this work is to provide an overview on the design and applications of context and context models for automation systems. A systematic literature review has been conducted, the results of which are represented as a knowledge graph., Comment: 10 pages, 5 figures, 1 table

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., Comment: 6 pages, 6 figures, 1 table. Accepted at CIRP ICME 2022