BMW Group (Germany)

Abstract Topology optimization is a key methodology in engineering design for finding efficient and robust structures. Due to the enormous size of the design space, evaluating all possible configurations is typically infeasible. In this work, we present an end-to-end, fault-tolerant quantum algorithm for topology optimization that operates on an exponentially large Hilbert space representing the design space. We demonstrate the algorithm on the two-dimensional Messerschmitt-Bölkow-Blohm beam problem. By restricting design variables to binary values, we reformulate the compliance minimization task as a combinatorial satisfiability problem, solved using Grover’s algorithm. Within Grover’s oracle, the compliance is computed through the finite-element method using established quantum algorithms, including block-encoding of the stiffness matrix, quantum singular value transformation for matrix inversion, Hadamard test, and quantum amplitude estimation. The complete algorithm is implemented and validated using classical quantum-circuit simulations. A detailed complexity analysis shows that the method evaluates the compliance of exponentially many structures in quantum superposition in polynomial time. In the global search, our approach maintains Grover’s quadratic speedup compared to classical unstructured search. Overall, the proposed quantum workflow demonstrates how quantum algorithms can advance the field of computational science and engineering.

The transition toward a Circular Economy (CE) has received significant attention from academia, industry, and policymakers; however, manufacturing practices remain predominantly linear, generating waste and inefficiencies. This study addresses the lack of accessible sustainability assessment methods by introducing the Circularity Calculator (CC), a novel tool for evaluating circular strategies during the early phases of process development. Unlike existing assessment frameworks, which often require extensive data and customization, the CC can be integrated directly to existing processes to combine environmental and economic impact into a streamlined evaluation process for early decision-making. The research involves collaboration with a leading German automotive manufacturer. Site visits and interviews enabled the identification of material flows and primary waste streams, which informed the definition of relevant indicators. The CC generates a dimensionless index, enabling comparison and prioritization of proposed scenarios without relying on supply-chain-wide data, which is often unavailable at early stages. Implications demonstrate the adaptability of the CC across industrial contexts, supporting conceptual planning and operational phases. Its intuitive design facilitates adoption by practitioners without extensive expertise in sustainability. The tool represents an advance in CE assessment, contributing to Sustainable Development Goals (SDGs) 9, 12, and 17 by promoting sustainable industrial practices, resource circularity, and collaborative evaluation frameworks.

Driving simulators are essential tools to guide automotive research and development. Their motion system requires a motion cueing algorithm (MCA) to keep the simulator motion platform within its physical boundaries, while simultaneously aiming to recreate the sensation of real vehicle motion. While traditional, filter-based approaches are still predominant, optimization-based MCAs have been at the center of MCA research for over a decade due to their ability to systematically improve motion cueing quality through explicit cost function design and constraints handling. However, despite their demonstrated advantages, these optimization-based methods have not yet achieved widespread adoption in driving simulation. This paper therefore provides a comprehensive review of optimization-based MCAs for driving simulation, categorizing and comparing algorithms, describing their key developments and core characteristics. The current limited real-time capability, lack of accurate evaluation methods, challenges in cost function design and its tuning, and the current lack of accurate future reference predictions are identified as key barriers to the practical deployment and widespread use of optimization-based MCAs. These theoretical and practical challenges are further reviewed, providing guidelines to advance the theory and application of optimization-based MCAs. Central in these advancements are a better understanding of which motions constitute a realistic motion experience, a framework allowing to compare the achieved motion fidelity of MCAs across papers, the design of the cost function focusing on human motion perception, and techniques for easing up the tuning process to swiftly reach high quality tunings for different simulators, scenarios, and use cases. We identify the need for improving the real-time capability, and providing high quality motion reference predictions using learning-based approaches on diverse datasets, along with techniques to handle existing uncertainties. Following these guidelines, new foundations for optimization-based algorithms in driving simulation can be achieved, which will significantly impact the research and development of automotive systems.

Perceiving the physical properties of different surfaces/textures via tactile sensing has been a long-standing problem in robotics. Most prior work has been limited to discriminative models that classify textures into a fixed set of categories. However, to enable seamless and efficient autonomous manipulation, robots must infer physical properties as structured, continuous variables rather than as discrete class labels. In this work, we present a novel deep state-space model (DSSM) to learn and infer key causal textural properties in an unsupervised manner. Using variational inference to solve the DSSM, our proposed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Latent Filter</i> allows robotic systems to perceive textures in a continuous and generalizable manner. In addition, we explore a novel interaction approach: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Tacser</i> (Tactile Enhancer), to further enhance tactile sensing through vibrations induced by high-frequency micro-movements and thereby improve perception. We evaluated our approach against state-of-the-art techniques and performed extensive ablation studies to demonstrate its effectiveness. This work advances tactile-based texture perception, providing a generalizable and comprehensive framework for robotics.

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;The reliability and durability of vehicles are crucial for the acceptance of new technologies by customers. Realistic test methods are necessary to validate or ensure the lifespan of vehicles and their components, particularly regarding specific conditions such as freeze start. This article provides an overview of the current state of research on the effects of freeze starts on the degradation of fuel cells. With this knowledge, relevant operating and boundary conditions for potential damage of the fuel cell are identified (e.g. start temperature, duration in subzero operation, dehydration). The field data from the BMW demonstrator fleet of iX5 Hydrogen Next were analyzed to gain insights into realistic freeze start related stress to the fuel cells. The dynamics of heating rates and the influence of the operating strategy are best represented on a Fuel Cell System (FCS). An experimental setup for a stack centered test on a FCS was developed including a climatic chamber and a subzero coolant supply in this study. The identified automotive conditions could be implemented similarly and reproducibly for the stack. In total of 140 freeze starts with the start stack temperatures between - 7 °C and - 18 °C were performed. These test results and the vehicle data were compared, and the limitations of this accelerated stress test are stated. The advantage of this method is the feasibility to validate the lifetime regarding freeze starts within a period of 1-2 month in 24/7 operation. The occurred problems during test development and their solutions are also described, and suggestions for improvement for less damaging freeze starts process are given.&lt;/div&gt;&lt;/div&gt;

The production process chain of lithium-ion battery (LiB) cells consists of several interconnected process steps with many parameters that can influence the final properties and quality of the cells. In the development of cylindrical LiB cells, steps such as winding, cell assembly, filling, and formation play an important role in determining the properties of the intermediate product and the final performance of the battery cell. Continuous and low scrap production in the experimental stages of LiB battery cell production is challenging due to the limited understanding of cause-and-effect relationships. For example, the machine parameters that need to be configured depending on the properties of the input material. As a result, individual tuning of production machine settings for different materials is required. This is particularly critical in research pilot production environments, such as the BMW Battery Cell Competence Centre, where different materials are tested and experimental approaches are required to optimize the manufacturing processes. This paper presents an approach for utilizing data analytics and machine learning with Open Platform Communications Unified Architecture (OPC UA) and Internet of Things (IoT) data pipeline to better understand the relationships between process parameters, input material, intermediate product features and final quality in the production of cylindrical LiB battery cells. In this study, different machine learning algorithms were tested to predict the initial setpoints of the routing path and the gripper values from the parameters of the winding process and the characteristics of the input material in order to obtain an intermediate product with the desired quality. The best results were obtained using Random Forest Regression with an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.976.

As touchscreen interfaces replace many physical controls in vehicles, safe and intuitive interaction becomes increasingly important. This study evaluated piezo-actuated haptic feedback versus conventional audio-only feedback across 11 use cases in parking and driving contexts. In a within-subject experiment with 23 participants in a fixed-based driving simulator, haptic feedback was rated as more satisfying, helpful, and supportive, particularly for scrolling or confirming actions while driving, despite showing no significant differences in objective performance measures. Participants described haptics as more engaging, though some noted habituation effects and inconsistencies. Audio-only feedback was seen as more stimulating but less consistently preferred. Findings suggest that contextual haptics improves perceived usability. Additionally, the study highlights the potential of semantic haptics—enabled by piezo technology—to convey meaning through expressive tactile patterns. Future research could explore adaptive multimodal systems and assess long-term effects under real-world conditions.

Purpose In the context of industrial production, the utilisation of data recording and processing techniques is becoming increasingly prevalent across the manufacturing sector. The solutions integrate sensors, facilitate the transmission of data, and enable data-driven decision-making, thereby reducing downtime and optimising quality. However, challenges emerge due to the limited non-transferable data or models between processes. Alterations to the production process can render collected data invalid, resulting in restricted datasets and potential overfitting. To address these issues, techniques such as data augmentation are employed. This study aims to develop a data augmentation methodology applicable in dynamic, data-scarce production environments, enhancing the robustness of regressor predictions. Design/methodology/approach The data augmentation method that has been developed is based on a sampling space that has been constructed on the grounds of the inherent uncertainty in the measured data on material properties. It is assumed that augmenting the data within the confines of its own uncertainty will ensure that the labels of the augmented data points can be preserved regarding their original counterparts. Findings The present study demonstrates, using data on car body part production in press shops, that the application of data augmentation techniques enhances the robustness of the models in question. Furthermore, the methodology developed has demonstrated superior performance compared to a standard augmentation technique, namely jittering. Originality/value The existing literature lacks data augmentation methods for regression tasks in manufacturing. This is because the majority of existing approaches were developed for classification purposes, given that data augmentation originated in image and sound processing.

• Driver fatigue in the context of CAD poses a significant safety concern, necessitating effective countermeasures. • A driving simulator experiment was conducted with 75 participants to investigate a countermeasure against driver fatigue. • While one participant group was allowed to engage in free-choice NDRAs during CAD, the other group was not allowed to engage in any activity. • Driver fatigue (measured with the Karolinska Sleepiness Scale) was significantly lower during CAD in the group with a free-choice NDRA than without; however, a free-choice NDRA did not positively affect driver fatigue and driving performance after CAD. • Findings suggest that free-choice NDRAs during CAD have a positive effect when the activity is performed but not beyond. A critical safety risk during conditional automated driving (CAD, SAE Level 3) is driver fatigue. Researchers have investigated several countermeasures to address this issue. Non-driving related activities (NDRAs) during CAD have shown considerable promise in preventing driver fatigue and should be evaluated in more detail. The aim of this study was to investigate free-choice NDRAs as a potential countermeasure against driver fatigue. For this purpose, 78 participants took part in a driving simulator experiment with a 2x2 mixed design. The participants drove manually before and after CAD to compare the impact of driver fatigue on driving performance. During CAD, the participants were randomly assigned to one of two conditions: with vs. without an NDRA. The participants of the first condition were allowed to engage in NDRAs after reaching a 7 on the Karolinska Sleepiness Scale (KSS) during CAD. Driver fatigue, takeover performance, and driving performance were assessed. The results revealed that driver fatigue was significantly lower during CAD in the group with an NDRA than without. However, the NDRA had no significant effect on driving performance (i.e., distance headway, time-to-collision, standard deviation of lateral position) in the second manual drive. Driver fatigue resulted in significantly worse driving performance in the second drive. The results showed that an NDRA only impacted driver fatigue during CAD, not manual driving performance afterwards.