Autonomous Driving is being utilized in various settings, including indoor areas such as industrial halls. Additionally, LIDAR sensors are currently popular due to their superior spatial resolution and accuracy compared to RADAR, as well as their robustness to varying lighting conditions compared to cameras. They enable precise and real-time perception of the surrounding environment. Several datasets for on-road scenarios such as KITTI or Waymo are publicly available. However, there is a notable lack of open-source datasets specifically designed for industrial hall scenarios, particularly for 3D LIDAR data. Furthermore, for industrial areas where vehicle platforms with omnidirectional drive are often used, 360° FOV LIDAR sensors are necessary to monitor all critical objects. Although high-resolution sensors would be optimal, mechanical LIDAR sensors with 360° FOV exhibit a significant price increase with increasing resolution. Most existing AI models for 3D object detection in point clouds are based on high-resolution LIDAR with many channels. This work aims to address these gaps by developing an automated AI-based labeling tool to generate 3D ground truth annotations for object detection from low-resolution LIDAR datasets captured in industrial hall scenarios. The point cloud data inside an industrial area at the KIT Campus Ost is recorded using a 16-channel LIDAR. The recorded objects include a forklift and box pallets for example. An upsampling LIDAR super-resolution approach is used that takes the recorded data as input for generating 64-channel point cloud data. The upsampled data is then utilized to fine-tune a 3D object detection model (Part-A2 net). Our testing results on a restricted dataset are highly promising, achieving a mean Average Precision of 95% at an IoU threshold of 0.75. The labeling tool is fully automated and utilizes the trained model for object detection. Manual corrections are also available. This research is part of the project FLOOW.

Target: Currently, too much humidity enters batteries through semipermeable membranes of emergency degassing, leading to unwanted corrosion. What methods can prevent this? Methodology: This is also a challenge for emergency degassing paired with a pressure equalization function. What materials and designs are appropriate to achieve the reduction? Approaches are: 1.) splitting the function into pure pressure equalization and into emergency degassing element without membrane. 2) reduction of the air flow rate of the membrane 3) Splitting of the membranes of the emergency degassing elements into gas-tight and semi-permeable. Results: The presentation looks at the different approaches in terms of technical feasibility, financial aspects and time required for production and installation. In addition, evaluations of tests are presented. The result is open-ended, each developer/manufacturer of batteries for BEV or NAT applications must decide for himself which approach is best for the application. Additionally, the aspects of PFOA and PFAS on the membrane material are considered. Discussion: How useful are the current approaches in terms of cost. Shouldn`t these aspects be incorporated earlier in the development and design of battery cases?

Slope differences and crosswind are decisive parameters for the energy consumption of vehicles. For electric vehicles in particular, both pieces of information are crucial for a precise determination of the remaining battery capacity and range. The team of authors from university and vehicle engineering contribute to the careful use of energy in the mobility of the future. Today, applications such as openweathermap.org/api or weatherapi.com or open-meteo.com/ provide the wind data for selectable geopositions. The presentation shows the conversion effort required to use this data and how the data can then be integrated into simulations and comparative measurements. The angle of slope also influences the energy consumption of the vehicle. Altitude data from GPS sensors or digital maps are often inaccurate and must be numerically post-processed for an exact calculation of the angle of inclination. For the first time, wind and altitude data recorded online are then integrated into driving simulations and evaluated to produce energy requirement forecasts for a target route. Documented journeys with a test vehicle and a comparison with longitudinal dynamics simulations round off the article. Janpoom, K. et al: Investigating the influential factors in real-world energy consumption of battery electric vehicles. Applied Energy 9 (2023) Daberkow, A. et al: An Energy Efficiency Comparison of Electric Vehicles for Rural–Urban Logistics. Springer, Small Electric Vehicles (2021)

This article deals with the creation and concretization of a scenario catalogue in the context of scenario-based validation of highly automated driving functions. Due to the possible combinatorics of the various driving situations and influencing parameters in the open-world context, it is not possible to simply execute all possible combinations. Instead, new methods are needed that can make a representative statement about the respective scenario, but at the same time attempt to reduce the degree of execution. To this end, a methodology for risk assessment and criticality analysis is to be developed as part of this work. This approach uses analogies to existing methodologies (e.g. ASIL) and transfers them to the scenario-based context. To this end, existing public data sets containing top-view camera recordings of real traffic scenarios were analysed and evaluated with regard to previously defined, scenario-specific parameters. Based on this, the methodology for the evaluation, risk assessment and prioritization of the specific scenarios in the context under consideration will be designed. Finally, the results will be categorized in the existing development cycle of scenario-based testing, implemented and evaluated and validated with the real system for an exemplary scenario.

Prof. Dr.-Ing. Hugo Gabele, LiMo 2040 Die Idee für ein modulares Fahrzeugkonzept, das (so gut wie) keinen Parkplatz mehr braucht und nachhaltig individuelle Mobilität mit urbanem Wohnen verbindet, wurde bereits auf dem Stuttgarter Symposium 2023 vorgestellt. Der Trick an der Sache ist die multifunktionale Kabine: Sie ist nicht nur Fahrzeug-Kabine, sondern auch Aufzugs-Kabine und vor allem Bestandteil einer modernen Wohnung. Neu ist die konkrete Umsetzung der Idee. Das Fahrzeugkonzept weist folgende Merkmale auf: Einfache und stabile Bauweise in U-Form (ähnlich DLR U-Shift). Kostengünstige konventionelle Lenk- und Antriebseinheit vorne (E-Motor mit Differenzial). Gefederte Einarmschwingen hinten. Platzsparendes Parken, U-Formen passen „gegenseitig“ zusammen. Das Carsharing-Prinzip reduziert zusätzlich den Parkplatzbedarf (auf 10 Kabinen kommen nur ca. 2 Fahrgestelle). Verzicht auf „Vollautonomes Fahren“ (dauert länger und wird teurer als von Experten prognostiziert) -> Teilautonomes System, d.h. Fahrgestell ausgerüstet mit GPS und Ultraschall-Sensoren (Autonomes Fahren nur vom ortsnahen Depot zum Gebäude). Teilautonomes bzw. manuelles Fahren zum Zielort mit Hilfe einer „Fernsteuerung“, die sich aus der Kabinenwand ausklappen lässt und die wichtigsten Fahrfunktionen incl. Joystick bzw. Lenkrad beinhaltet. Anfang 2024 wird der Prototyp in Betrieb genommen und ggf. erstmals auf dem Stuttgarter Symposium vorgestellt.

AURIX TC2xx and TC3xx microcontrollers are used in numerous automotive applications, and recently, also in some which include machine learning tasks. Yet, these applications are mainly engineered manually, and only little tool support exists for bringing neural networks to TriCore microcontrollers. Thus, we propose OpTC, an end-to-end toolchain for automatic conversion, code generation, and deployment of neural networks on TC3xx microcontrollers. OpTC supports various types of neural networks and provides compression (e.g., pruning) and optimization (e.g., approximation of special activation functions) techniques often required to reduce execution time and memory footprints while maintaining accuracy. The toolchain incorporates cost models for estimating memory utilization and execution time of a neural network model. The flexibility in supporting different types of neural networks, such as multi-layer perceptrons (MLP), convolutional neural networks (CNN) and recurrent neural networks (RNN), is shown in case studies for a TC387 microcontroller. Automotive applications for predicting the temperature in electric motors and detecting anomalies are thereby used to demonstrate the effectiveness and the wide range of applications supported by OpTC.

Previous studies have shown that dosing AdBlue into the exhaust system of diesel engines to reduce nitrogen oxides can lead to an increase in the number of particles (PN). In addition to the influencing factors of exhaust gas temperature, exhaust gas mass flow and dosing quantity, the dosed medium itself (AdBlue) is not considered as a possible influence due to its regulation in ISO standard 22241. However, as the standard specifies limit value ranges for the individual regulated properties and components for newly sold AdBlue, in reality there is still some margin in the composition. This paper investigates the particle number increase due to AdBlue dosing using several CPCs. The increase in PN is determined by measuring the number of particles after DPF and thus directly before dosing as well as tailpipe. Several AdBlue products from different sources and countries are measured and their composition is also analyzed with regard to the limit values regulated in the standard. This shows that differences in the PN increase can be determined for the various products. In addition, two measurements are carried out with pure water as a main component of AdBlue in the form of single and double distilled water. Interestingly, the dosing of pure water also shows an increase in PN depending on the purity of the water. Furthermore, two AdBlue products are artificially aged in order to violate the standardized limit values, which is a feasible use case with regard to ISC tests, and subsequently measured. Since these impurities cannot be influenced but have a noticeable effect on the measured PN, it is important to quantify this and, if necessary, to take it into account in legislation.

Due to manifold benefits compared to proprietary software solutions, free and open source software (FOSS) in general, and Linux especially becomes more and more relevant for embedded solutions in the automotive domain, especially in High Performance Computing Platforms (HPC). However, taking over liability and warranty for a FOSS software-based problem raises the problem of software quality assurance, and thus respectively risk control. In order to control and minimize the residual risk of a product or service, the traditional and well-accepted measure in the automotive domain is to assess the engineering processes and resulting work products via a process assessment model given by the ASPICE maturity model, as well as requirements from functional safety standards for safety related functions. The underlying process reference model of ASPICE assumes software development performed and controlled by an organization. However, this assumption is not given by and even contrary to the nature of FOSS development, where high quality is achieved based on feedback and contributions of an open community. While typical software quality assurance measures are widespread in community-based software development, a single entity cannot control these. This, along with the huge code base in Linux makes applying the low-level software related processes ASPICE Process Reference Model (PRM) both meaningless and economically infeasible. In this paper, we propose a tailoring of the process model accompanied with compensation measures, which accounts for the FOSS software specifics. This allows to achieve the quality assurance and risk mitigation goals of ASPICE, and consequently an assessment via the ASPICE Process Assessment Model (PAM) as well as functional safety standards. We further provide details on our solutions and strategies to fulfill the key elements of the process model. The solution presented here is one key factor for our EB corbos Linux – built on Ubuntu to provide a production grade Linux distribution suited to the automotive embedded needs, including liability, warranty, and long-term maintenance.

By means of a system analysis of heavy electric trucks that are on the market or about to be introduced with a permissible total weight ≥ 12 tons, the drive structures known today are analyzed and structured. Hence the changes compared to today`s internal combustion engine layout are shown. The advantages and disadvantages with regard to the integration and function of the battery-electric powertrains are then elaborated. Based on an analysis and compilation of the main areas of application in distribution transport, a ranking of the electric truck system structures is carried out with the perspective of universal and commercially value-adding use in these applications. With the knowledge gained about the integration of the electric drive and high efficiency in operation, a concept for an ideal structure of an electric truck for these applications is described and its layout is compared with the internal combustion engine system structure

The thermal management of electric vehicles is becoming increasingly important for achieving future development goals. In particular, the automotive industry is focusing on increasing vehicle efficiency by utilizing existing heat quantities in conjunction with the use of heat pump systems. This has a direct impact on the design and layout of thermal systems and the respective components. This publication presents the development of a simulative development platform for the design of thermal system components. In particular, differences between current development approach and approach utilizing new simulation environment will be explained. The development steps required to create the necessary sub models and their coupling will then be discussed. In an initial investigation, modern electric vehicles operating points and their influence on cooling module and heat exchanger design are considered. Comparison of selected results from new simulation platform and from original development approach is processed and discussed for new areas of improvement for system and components. Performance and durability-related data are collected to describe modern electric vehicle configuration impact on thermal management system. The article concludes with an outlook on how the development platform can be further improved and used for the development of thermal system components in the future.

This work explores the methanol compression ignition combustion assisted with glow plug in a light-duty diesel engine. An extensive computational study was conducted to optimize the position of the glow plug. The effects of spray umbrella angle, the relative angle between the glow plug and jet, and the injection strategy on the engine performance were evaluated. Of these parameters, the relative angle between the glow plug and jet was found to be the dominant factor affecting the ignition and flame propagation processes. At each position of the glow plug, the optimum relative angle differed due to the complex flow and air-fuel mixing within the combustion chamber. The removal of the swirling flow resulted in the earlier ignition and combustion phasing. However, the engine`s thermal efficiency was reduced due to the increased combustion loss. In addition, compared to the single injection strategy, the split injection strategy was more effective in promoting the ignition process. The narrower spray included angle yielded better fuel economy because less fuel was trapped within the squish region, leading to faster flame propagation and more advanced combustion phasing. However, even after optimizations, the engine still suffered from a high incomplete combustion loss. To mitigate this issue, the effect of dual glow plugs was examined. The engine performance was significantly improved due to the generated multi-ignition pockets.

At the German Aerospace Center (DLR), scientists in the field of transportation research are working on new transport and vehicle concepts to create solutions for sustainable and user-friendly mobility. The DLR Institute of Vehicle Concepts Stuttgart (FK) has developed a concept study for a modular and fully automated transportation system, the U-Shift concept. The special feature of the U-Shift concept is the separation between the driveboard, the drive unit with drivetrain, automation system and lifting mechanism and different vehicle superstructures, the capsules. These capsules can be designed in different shapes and transport people or goods. The capsules are changed using a lifting mechanism integrated into the drive board. There are now a large number of U-Shift projects with different focal points and participating companies or research institutions such as DLR, KIT, FKFS or the University of Ulm. For the submitted paper, project contents from U-Shift IV / BUGA 23 are to be described. A new research demonstrator was developed by DLR and project participants from industry with a driveboard, passenger capsule and multi-use platform for operation at the Federal Garden Show in Mannheim 2023. The funding was supported by the Strategy Dialogue for the Automotive Industry Baden-Württemberg (SDA) with a focus on technology transfer.

The verification, validation and homologation of automated driving systems is highly dependent on the HD-Map availability for the ODD creation. Currently this is the bottleneck limiting the possibility of virtual verification and validation on available HD maps which are predominantly non-urban. Mapping areas conventionally is both costly and time consuming.To address this challenge, AVES Reality and AVL have joined forces. AVES offers an automated HD- Maps generation process and creates 3D environments from aerial images anywhere on the globe. Additionally, its parameter-based semantic environment reconstruction allows customization and variation to match ODD requirements. This enables early-stage assessment and evaluation of new areas for AD-Systems also in urban areas, long before detailed HD-Maps are created. AVL leverages these HD-Maps and 3D environments by integrating them in an AVL simulation toolchain and provides the customers with the possibility to run hundreds of thousands of scenarios per day frenabling V&V and SOTIF testing activities for automated driving. In summary, AVES Reality and AVL accelerate this process and enable the scaling of V&V and homologation of automated driving systems. The poster will introduce in the challenges and requirements of scenario-based testing using simulation for automated driving, the scope of the collaboration between AVL and AVES Reality and the solution approach we provide to enable and improve AD simulation.

Due to its physical and chemical properties, hydrogen is an attractive fuel for internal combustion engines, providing grounds for studies on hydrogen engines. It is common practice to use a mathematical model for basic engine design and an essential part of this is the simulation of the combustion cycle, which is the subject of the work presented here. One of the most widely used models for describing combustion in gasoline and diesel engines is the Wiebe model. However, for cases of hydrogen combustion in DI engines, which are characterized by mixture stratification and in some cases significant incomplete combustion, practically no data can be found in the literature on the application of the Wiebe model. Based on Wiebe`s formulas, a mathematical model of hydrogen combustion has been developed. The model allows making computations for both DI and PFI hydrogen engines. The parameters of the Wiebe model were assessed for three different engines in a total of 26 operating modes. The modified base model considers the significant incompleteness of hydrogen combustion, which occurs at high air/fuel equivalence ratio. For PFI and DI hydrogen engines, equations and numerical values for the Wiebe model coefficients were determined to describe the dynamic and duration of combustion. Based on our simulation results we suggest using the sum of two Wiebe curves to describe combustion in zones with a lean mixture in DI engines. This allows a more accurate characterization of the combustion dynamics and pressure curves. In order to model a double hydrogen injection, we suggest using the sum of three Wiebe curves representing the combustion of the first injection in the flame front, the diffusion combustion of the second injection, and the relatively slow combustion in lean mixture zones. In the paper, we present a method for selecting the coefficients of each of the Wiebe curves.

The European Green Deal aims to reduce emissions by 55% compared to 1990 levels by 2030. The transportation sector has lagged behind in reducing emissions since 1990, and to address this, it must electrify powertrains, transform the mobility model, and decarbonize existing vehicle stock. Electrification of powertrains, including hybrids and battery electric vehicles, is the primary driver for decarbonizing passenger car transport, while hydrogen fuel cells will play a complementary role, particularly in commercial vehicle transport. To achieve this, battery manufacturing capacity must be increased by 33% annually to reach 5.8TWh by 2030 and transformed from linear to circular. Additionally, the charging infrastructure must be expanded 3-7x faster than EU`s weekly addition of 2000 new charging points today to meet the expected demand. Bio- and e-fuels are another approach to reducing the carbon dioxide emissions of existing internal combustion engine vehicles. Alternatives to privately owned passenger vehicles, such as public transit and micromobility options, are insufficient to drive the necessary shift in demand away from private cars. Therefore, the implementation of pooled shared transport, or "ridesharing," on a large scale is the next crucial step in reducing the proportion of privately-owned vehicles in passenger mobility, with anticipated demand reaching hundreds of thousands of vehicles across the top 30 cities in the European Union by 2030.

**Presentation only in German language** Open-pore aluminium chill casting is an established, sustainable series technology. The new class of materials with new mechanical, thermal, acoustic and other properties defines the next evolutionary stage in the development of new types of thermal components and systems with many new potentials that cannot be realised with other technologies. Hundreds of possible material variants, new phenomena of heat and mass transfer, novel mono-material hybrids and multi-material systems, modified material properties and previously unthinkable design possibilities form a completely new world for strategic product development. The world`s first feasibility study will be presented. It showing the substitution of a conventional heat sink for power electronics with a system that is up to 70% smaller and up to 70% lighter. There is also an outlook on new potential for heat exchangers, heat accumulators, thermally conductive plastic parts and other components. Customer feedbacks "The idea of having a heat sink the size of a bar of chocolate is really very tempting. We want to order samples“. / CEO, Manufacturer power electronics. „We have completed the tests. The results are promising“. / R&D Manufacturer power electronics. "We see an opportunity to strategically open up new markets with our products" / CTO, Heat exchanger manufacturer. www.porecool.com

The increasing complexity and diversity of modern vehicles requires efficient development methods and processes to ensure the profitability of a vehicle manufacturer in the competitive environment. Therefore, this paper presents a generic approach and related methods for the consistent virtual development of combined longitudinal and lateral vehicle dynamics. This process is based on the established V-model from systems engineering and is exemplary applied for the development of a mid-size electric SUV. First step of the approach is the definition of vehicle dynamics targets based on objective evaluation criteria. Afterwards, these targets are translated into target ranges of the relevant systems and components parameters of the vehicle using an automated target cascading process. To support the calibration of vehicle dynamics control systems, a virtual calibration method involving sensitivity analysis and optimization techniques is proposed. The application of the methods for the development of a sample vehicle demonstrates their effectiveness and the potential for enhancing the efficiency of the vehicle development process.

As the technology in cars continues to evolve, the development is faced with increasingly complex software solutions. Tasks such as maintenance over the lifespan of the vehicle, reuse in other models as well as compliance with regulatory requirements, therefore, need to be efficiently accomplished to keep up with ever shortening development cycles. In the context of the research project SmartDelta (sponsored by the Federal Ministry of Education and Research on the basis of a decision by the German Bundestag via ITEA4), we are developing a tool for code-based similarity analysis. This enables us to obtain automated code reuse suggestions, which makes code changes such as fixing vulnerabilities over multiple product variants faster. Combined with test case prioritization, it will significantly decrease time needed for testing. By taking advantage of the collaboration possibilities within the project, our solution is created with large, production scale codebases in mind. Our approach is based on Code Property Graphs which provide a compact yet comprehensive model on the code. We employ classical algorithms in conjunction with machine learning to extract the desired information out of Code Property Graphs.

In electrified vehicles, auxiliary units can be a dominant source of noise, one of which is the re-frigerant scroll compressor. Compared to vehicles with combustion engines, e-vehicles require larger refrigerant compressors, as in addition to the interior, also the battery and the electric motors have to be cooled. Currently, scroll compressors are widely used in the automotive industry, which generate one pressure pulse per revolution due to their discontinuous compression principle. This results in speed-dependent pressure fluctuations as well as higher-harmonic pulsations that arise from reflections. These fluctuations spread through the refrigeration circuit and cause the vibration excitation of refrigerant lines and heat exchangers. The sound transmission path in the air con-ditioning heat exchanger integrated in the dashboard is particularly critical. Various silencer con-figurations can be used to dampen these pulsations. This paper compares the acoustic damping performance and pressure loss of two mufflers and a resonator for different operating points. Measurements of the pressure pulsations before and after the flow silencer are carried out using a refrigeration circuit acoustic test rig. The experimentally determined values are compared with analytically calculated sound attenuation curves of the silencers. In addition, the pressure loss generated by the flow silencer is measured, as pressure losses in the refrigeration circuit have a negative effect on thermal efficiency. The damping range of the resonator starts at a lower frequency compared to the mufflers. This allows the reduction of high-amplitude 1st order pressure pulsations. Furthermore the resonator generates a lower pressure loss than the mufflers. The findings on the operation principle and damping performance of different refrigerant circuit silencers enable the reduction of flow-induced noise in thermomanagement system components in vehicles.

This paper proposes a novel approach to the design of a Hardware Abstraction Layer (HAL) specifically tailored to embedded systems, placing a significant emphasis on time-controlled hardware access. The general concept and utilization of a HAL in industrial projects are widespread, serving as a well-established method in embedded systems development. HALs enhance application software portability, simplify underlying hardware usage by abstracting its inherent complexity and reduce overall development costs through software reusability. Beyond these established advantages, this paper introduces a conceptual framework that addresses critical challenges related to debugging and mitigates input-related problems often encountered in embedded systems. This becomes particularly pertinent in the automotive context, where the intricate operational environment of embedded systems demands robust solutions. The HAL design presented in this paper mitigates these issues. The design is structured as a modular software concept, leveraging the strategic use of configuration tables to provide an abstracted, rapid and well-organized method for configuring hardware. Furthermore, those configuration tables are used to realize an application-specific time-controlled synchronization mechanism between the actual hardware data registers and an internal software representation of those. The application software exclusively interacts with this representation, preventing errors arising from unstable inputs and ensuring strict timing. This paper provides a detailed description of the design, with a focus on its modular structure for an efficient and memory-saving implementation. Moreover, the document explores and discusses potential extensions and adaptations to the proposed design, enhancing its flexibility for individual use cases. In conclusion, this comprehensive exploration seeks to contribute to the advancement of embedded systems development by offering a refined and adaptable HAL design.

2023 witnessed a surge in cybersecurity incidents within the automotive sector: major US automotive supplier victim to ransomware, Porsche Macan sales halting due to cybersecurity non-compliance, recurring hacker intrusions into infotainment systems at Tesla, Hyundai, Ford, and others. The trend continues into 2024, with a new focal point: the applicability of UN R155 and R156 to all vehicles, encompassing not only new models but also newly manufactured former models. Cybersecurity is evolving beyond a safety or operational concern for OEMs; it is now a compliance risk that demands attention. One essential pillar of regulation consists of managing risks adequately, timely and efficiently. Adequately, because the risks must be identified and assessed, including the complexity due to functions and components interactions. Timely, the risks must be continuously re-evaluated before and after production, to follow software changes and to adapt to the threat landscape. Efficiently, because risks mitigated via security controls must be tracked down along the development process, to ensure good implementation and alignment between theoretical artefacts and real residual in-vehicle risks. Unfortunately, many OEMs are not ready to deal with it. Poor methodologies associated with outdated tooling jeopardize the whole security chain. In this paper, we will discuss pitfalls, upcoming challenges for stakeholders in 2024 and how to build viable solutions for efficient vehicle security.

The inclusion of simulation-based or simulation-supported validation can advance the validation process in terms of reducing testing time and the required number of samples. In this presentation and paper a generic advanced model based Design Validation Plan (DVP) for e-axles is introduced. It is shown how simulation can be used to substitute physical testing for design validation and save about 50% of time and cost efforts compared to a generic common European standard validation process. One common testing area with a high potential for virtual validation is the assessment of efficiency. Latest results of an AVL research project regarding an e-motor efficiency mapping are shown and discussed. Focus is the correlation between simulation and measurement and in particular the investigation and comparison of validation confidence including uncertainties of both, virtual and physical validation. These investigations underline the trustworthiness of simulation – which is key to implementing advanced model based virtual validation to the industry.

The future success of the automotive OEM will closely be related to the adaption of artificial intelligence (AI) in all its domains, e.g., in the engineering process, in production, and in the software-defined vehicle itself. In more detail, fast AI surrogate models can speed up time-consuming simulations in the engineering process by a huge factor – independent of the simulation domain, being it fluid, structural, thermal, or electrical. Hence, AI surrogate models can be used to identify the optimal configuration, something that is often practically impossible using traditional simulations. Moreover, the automotive OEM owns huge datasets, originating for example from the development and testing, from production or from the fleet. To gain knowledge from these datasets can be of upmost importance. One key is the data-driven detection of unexpected data using AI, for example, for predictive maintenance or for AI-driven data correction. In addition, one can use data to predict its future evolution to predict, e.g., driving profiles or the power consumption of the vehicle electrical system. In this talk, we introduce the AI framework OPTIMALIS® , demonstrate its broad range of application and highlight its intuitive usage. The key advantages of OPTIMALIS® are its standardized data-interface, its ability to automatically adopt to data, and its autonomy. As such, OPTIMALIS® is the ideal tool to test the usage of AI in your use-case and to leverage your business to the next level.