Daimler Truck (Germany)

Serious games have become an important tool to train individuals in a range of different skills. Importantly, serious games or gamified scenarios allow for simulating realistic time-critical situations to train and also assess individual performance. In this context, determining the user's cognitive load during (game-based) training seems crucial for predicting performance and potential adaptation of the training environment to improve training effectiveness. Therefore, it is important to identify in-game metrics sensitive to users' cognitive load. According to Barrouillets' time-based resource-sharing model, particularly relevant for measuring cognitive load in time-critical situations, cognitive load does not depend solely on the complexity of actions but also on temporal aspects of a given task. In this study, we applied this idea to the context of a serious game by proposing in-game metrics for workload prediction that reflect a relation between the time during which participants' attention is captured and the total time available for the task at hand. We used an emergency simulation serious game requiring management of time-critical situations. Forty-seven participants completed the emergency simulation and rated their workload using the NASA-TLX questionnaire. Results indicated that the proposed in-game metrics yielded significant associations both with subjective workload measures as well as with gaming performance. Moreover, we observed that a prediction model based solely on data from the first minutes of the gameplay predicted overall gaming performance with a classification accuracy significantly above chance level and not significantly different from a model based on subjective workload ratings. These results imply that in-game metrics may qualify for a real-time adaptation of a game-based learning environment.

Abstract Computerized systems are taking on increasingly complex tasks. Consequently, monitoring automated computerized systems is becoming increasingly demanding for human operators, which is particularly relevant in time-critical situations. A possible solution might be adapting human–computer interfaces (HCI) to the operators’ cognitive load. Here, we present a novel approach for theory-based measurement of cognitive load based on tracking eye movements of 42 participants while playing a serious game simulating time-critical situations that required resource management at different levels of difficulty. Gaze data was collected within narrow time periods, calculated based on log data interpreted in the light of the time-based resource-sharing model. Our results indicated that eye fixation frequency, saccadic rate, and pupil diameter significantly predicted task difficulty, while performance was best predicted by eye fixation frequency. Subjectively perceived cognitive load was significantly associated with the rate of microsaccades. Moreover our results indicated that more successful players tended to use breaks in gameplay to actively monitor the scene, while players who use these times to rest are more likely to fail the level. The presented approach seems promising for measuring cognitive load in realistic situations, considering adaptation of HCI.

Autonomous Vehicles (AVs) are supposed to be continuously updated, besides that they combine many diverse applications by various developers. Accordingly, they require flexible and dynamic software (SW) architectures. With communication paradigms changing from signal-based to service- oriented architectures (SOAs), long-established standards seem no longer sufficient. The AUTomotive Open System ARchi-tecture (AUTOSAR) Platform came up with the Adaptive standard to meet future automobile requirements. Alongside, while requirements of AV s and robotic systems are similar to a certain extent, the Robot Operating System (ROS) gets growing attention in the automotive environment. With the introduction of ROS2 and ROS2-based commercial solutions, the Adaptive AUTOSAR Platform seems not to be the only appropriate standard for developing high-performing SW architectures. This paper provides an overall picture of Adaptive AUTOSAR and ROS2 and evaluates ROS2 and its suitability in an auto-motive context based on the Adaptive AUTOSAR architecture. This research analyzes the extent to which ROS2 fulfills the functionality provided by Adaptive AUTOSAR.

Growing numbers of plug-in electric vehicles in Europe will have an increasing impact on the electricity system. Using the agent-based simulation model PowerACE for ten electricity markets in Central Europe, we analyze how different charging strategies impact price levels and production- as well as consumption-based carbon emissions in France and Germany. The applied smart charging strategies consider spot market prices and/or real-time production from renewable energy sources. While total European carbon emissions do not change significantly in response to the charging strategy due to the comparatively small energy consumption of the electric vehicle fleet, our results show that all smart charging strategies reduce price levels on the spot market and lower total curtailment of renewables. Here, charging processes optimized according to hourly prices have the strongest effect. Furthermore, smart charging strategies reduce electricity purchasing costs for aggregators by about 10% compared to uncontrolled charging. In addition, the strategies allow aggregators to communicate near-zero allocated emissions for charging vehicles. An aggregator's charging strategy expanding classic electricity cost minimization by limiting total national PEV demand to 10% of available electricity production from renewable energy sources leads to the most favorable results in both metrics, purchasing costs and allocated emissions. Finally, aggregators and plug-in electric vehicle owners would benefit from the availability of national, real-time Guarantees of Origin and the respective scarcity signals for renewable production. • Agent-based simulation of ten European electricity markets in Central Europe. • Simulating millions of EV under four different smart charging strategies by 2030. • Impacts on prices, curtailment, production- and consumption-based emissions. • All charging strategies reduce spot market prices and total renewable curtailment. • Charging with renewables in real-time minimizes purchasing costs for aggregators.

Subcooled liquid hydrogen (sLH2) is an onboard storage, as well as a hydrogen refueling technology that is currently being developed by Daimler Truck and Linde to boost the mileage of heavy-duty trucks, while also improving performance and reducing the complexity of hydrogen refueling stations. In this article, the key technical aspects, advantages, challenges and future developments of sLH2 at vehicle and infrastructure levels will be explored and highlighted.

Durability is a major lever for commercial success of proton exchange membrane fuel cells (PEMFCs). The introduction of OER catalyst to the PEMFC anode has been established as a material based mitigation strategy for reversal events caused by gross fuel (i.e. H 2 ) starvation. We investigated the degradation of two different OER based reversal tolerant anodes during short-term recurring reversal operation to mimic field occurrence of reversal events realistically. PEMFC failure during normal operation can be observed whereas OER activity during reversal operation is unaffected. This result is in contrast to findings for commonly applied prolonged reversal accelerated stress tests (ASTs) and indicates an OER catalyst recovery effect for short and recurring reversal events. Combining the developed AST with cyclic voltammetry, electrochemical impedance spectroscopy and hydrogen pump, tests failures during normal operation is mainly assigned to hydrogen oxidation mass transfer increase indicating carbon corrosion and structural change within the anode catalyst layer. Consequently, the developed combination of AST and further characterization methods enables in situ distinction between catalyst and structural degradation, highlighting to be a good basis to investigate future aspects regarding anode degradation caused by cell reversal.

Pedestrian tracking in advanced driver assistance systems in commercial vehicles is not only important in the frontal field of perception, but also in the blind spot of the vehicle (right side), e.g., to mitigate or avoid collisions during turning maneuvers. While a camera system and radars observe the front, only a radar is available at the vehicle's side. This paper will present a Joint Integrated Probabilistic Data Association Filter (JIPDAF) that tracks pedestrians in the frontal field of view and in the vehicle's blind spot. Although the sensors do not have a common field of view, we show that tracking across the blind region is advantageous, since information that has already been retrieved by the front sensors can be conserved, and the confirmation time of the tracks could be reduced. The results include a comparison of the JIPDAF approach running in real-time with an extended Kalman Filter with global nearest neighbor data association using data from real measurements. Furthermore, we will compare the fusion results to measurements of a 3D laser scanner. To the authors' knowledge, there is no JIPDAF approach for pedestrian tracking using camera HOG detections and radar sensors yet.

This paper focuses on the steps to compare methods to detects faults at the end of line (EOL) testing section after the assembly of permanent magnet synchronous motors (PMSM) for automotive, especially truck applications. In detail two philosophies were discovered in case of using end of line test benches to check all functionalities of an electric motor after all assembly steps. The comparison is a practical approach and involves three parts to investigate on the passive end of line test bench as a sufficient method to detect all critical faults and the performance of PMSMs.

A Vibrating Intrinsic Reverberation Chamber (VIRC) has been installed in a semi-anechoic chamber that is normally used for electromagnetic interference test in the automotive industry and its performances evaluated. This hybrid strategy permits converting a traditional semi-anechoic chamber into a reverberating environment in a few hours. Furthermore, the proposed solution makes possible the use of two different types of testing environment without searching for additional space in the testing house. The work analyzes some figure of merits that are normally used for traditional reverberation chambers like the autocorrelation of the electromagnetic field and its statistical uniformity and isotropy properties. The analysis is performed in the frequency range from 100 MHz to 6 GHz in a 26 m x 8 m x 6 m cavity. The experimental datasets are finally tested through a goodness-of-fit test for checking their convergence to theoretical values.

A heavy-duty engine is type-approved in engine dynamometers, while its in-service conformity is controlled on the road. In the first case, laboratory particle number systems (LABS) sample from a full dilution tunnel or a proportional partial flow dilution system (PFDS). In the second case portable emissions measurements systems (PEMS) measure directly from the tailpipe. Permitting in the regulation LABS sampling directly from the tailpipe would simplify testing and would improve their comparability with PEMS. In this study PEMS and LABS, both sampling from the tailpipe and measuring solid particles >10 nm, were compared with references systems (i.e. LABS from PFDS). One compressed natural gas (CNG) engine, and three diesel engines, all Euro VI step E, with or without urea injection, and with or without crankcase ventilation connected to the tailpipe, challenged the systems with different emission levels and particle sizes and properties. The results showed that the differences of the LABS to the references were in most cases within ±25%, with a few exceptions. The PEMS were within ±50%. There was no or small effect on the differences from engine technology, urea injection or crankcase ventilation. The inclusion of sub-23 nm particles increased 100% to 250% the particle number emissions. The urea injection increased the >10 nm emissions 300–600% (2–5 × 1010 p/kWh). Connecting the crankcase ventilation to the tailpipe further increased the >10 nm particle number emissions 340–560% (1.4–2.5 × 1011 p/kWh), bringing the >10 nm levels of the engines to approximately half of the current particle number limit, applicable to particles >23 nm.

Classification techniques are increasingly adopted for quality control in manufacturing, e.g., to help domain experts identify the cause of quality issues of defective products. However, real-world data often imply a set of analytical challenges, which lead to a reduced classification performance. Major challenges are a high degree of multi-class imbalance within data and a heterogeneous feature space that arises from the variety of underlying products. This paper considers such a challenging use case in the area of End-of-Line testing, i.e., the final functional test of complex products. Existing solutions to classification or data pre-processing only address individual analytical challenges in isolation. We propose a novel classification system that explicitly addresses both challenges of multi-class imbalance and a heterogeneous feature space together. As main contribution, this system exploits domain knowledge to systematically prepare the training data. Based on an experimental evaluation on real-world data, we show that our classification system outperforms any other classification technique in terms of accuracy. Furthermore, we can reduce the amount of rework required to solve a quality issue of a product.

Current development of advanced driver assistance systems (ADAS), e.g., for collision mitigation are increasingly concerned about the protection of other road users. Environment perception provides objects like cars or pedestrians. A prediction of the system vehicle's path is required to decide about the relevance of the objects for a system reaction or to reduce the CAN load in advance. A standard measure for the object criticality is the time-to-collision, where the system vehicle's path is predicted under the assumption of constant acceleration and yaw rate or using lane markings. Lane markings often are not available on urban streets, and the vehicles do not necessarily follow the own lane, e.g., due to parked cars at the road side. This paper proposes an approach that uses maneuver classification based on a combination of the longest-common-subsequence method and a Bayesian classifier. The knowledge obtained about the maneuver in the classification step is used to predict the future trajectory in a parameterizable way. The approach is evaluated in comparison to a prediction with constant acceleration and constant yaw rate using recorded data from more than 20 hours of driving.

Abstract With the introduction of advanced driving assistance systems managing longitudinal and lateral control, conditional automated driving is seemingly in near future of series vehicles. While take-over behavior in the passenger car context has been investigated intensively in recent years, publications on semi-trucks with professional drivers are sparse. The effects influencing expert drivers during take-overs in this context lack thorough investigation and are required to design systems that facilitate safe take-overs. While multiple findings seem to cohere in passenger cars and semi-trucks, these findings rely on simulated studies without taking environments as found in the real world into account. A test track study was conducted, simulating highway driving with 27 professional non-affiliated truck drivers. The participants drove an automated Level 3 semi-truck while a non-driving-related task was available. Multiple time critical take-over situations were initiated during the drives to investigate four main objectives regarding driver behavior. (1) With these results, comparison of reaction times and behavior can be drawn to previous simulator studies. The effect of situation criticality (2) and training (3) of take-over situations is investigated. (4) The influence of warning expectation on driver behavior is explored. Results obtained displayed very quick time to hands on steering and time to first reaction all under 2.4 s. Highly critical situations generate very quick reaction times M = 0.81 s, while the manipulation of expectancy yielded no significant variation in reaction times. These reaction times serve as a reference of what can be expected from drivers under optimal take-over conditions, with quick reactions at high speed in critical situations.

Battery packs consist of multiple single cells, which are connected in series. During service life the performance parameters of each cell behave individually under constant environmental conditions causing significant differences. These differences are attributed to variances in process and product parameters within the production line of Li-ion cells and directly affect the performance of battery packs, in which the cell with the steepest aging curve, referred as the worst-case cell, defines the overall performance of the battery. As a result of accelerated aging tests, with a focus on enhancing the effect of lithium plating, we show, that not only the cells have individual aging behavior through their life cycle, but also quantify noticeably growing capacity spread in the cell compound towards end of life and show the impact on the use case of battery systems during lifetime. A mathematical model is introduced to fit & analyze the degradation curve of 50 tested lithium-ion cells. This novel model allows, compared to normally used regression curves, to gain information about how certain aging mechanisms affect the capacity degradation & quantifies the capacity loss related to these. As a result, a comparison of the effect of these aging mechanisms between different cell chemistries is possible. Additionally, the quality of cell production lines and the impact on the applications can be evaluated by using the model. Furthermore, the visualization of characteristic areas within the life cycle of a battery cell is possible & gives useful information regarding the decrease of the impact of certain aging mechanism during the development phase or suitable SoH strategies during life cycle.

The topic of fast charging lithium-ion batteries is of great importance, especially with regard to e-mobility. Thereby, fast charging while retaining a long life-time in particular is of special significance. The common charging methods are usually based on a continuous charging current. In this work, a different type of charging process is considered – the so-called pulse charging. The focus is set on high-frequency pulse charging profiles. A variation of the parameters frequency, duty cycle, amplitude current as well as the end-of-charge voltage is investigated and their influence on a commercial high-energy cylindrical cell in 21,700 format is determined. Profiles with a frequency of 100 Hz, 500 Hz and 1 kHz and a duty cycle of 50 %, 70 % and 90 % are analyzed. In addition, end-of-charge voltages from 4.2 V up to 4.4 V are investigated. The continuous current (CC) charging process is used as a reference. A cycle test is also performed with a promising pulse charging profile. The results indicate that the investigated pulse charging profiles cannot be recommended as an innovative charging method. It is true that an acceleration of the charging process is achieved. However, this is due to the fact that less capacity is charged into the cell. Significantly more capacity can be charged into the cell by increasing the end-of-charge voltage. However, lithium plating may occur and there is a strong decrease in capacity during the cycle test. • High-frequency pulse charging of state-of-the-art lithium-ion cell was investigated. • Investigated pulse charging profiles have no positive effect compared to common methods within common end-of-charge voltage. • It's true that acceleration of the charging time can be achieved, but this is due to less capacity is charged into the cell. • High capacities would be achievable by increasing the end-of-charge voltage but this leads to accelerated aging of the cell.

Platooning is a cooperative driving technology for driving in a longitudinal formation. Vehicle-to-vehicle (V2V) communication is the key enabling technology. Platooning is considered to be a closed application between well-known participants that use V2V data for their operation. For the deployment of platooning, multi-brand, i.e. interoperable, solutions are paramount. To achieve this, the European ENSEMBLE (ENabling SafE Multi-Brand pLatooning for Europe) project has been conducted. In this project, multi-brand specifications have been harmonized to a set of common specifications and requirements for a generic Platooning Support Function (PSF). Part of the specifications is a common interaction protocol, which also encompasses security features. A reference design has been developed to initially check the specifications. It consists of a detailed design of the common functionality and includes software and communication sub-system hardware prototypes. This reference design has also been used as benchmark for brand-specific implementations, e.g. in communication plug tests. Daimler Truck included the reference design directly in their PSF implementation, which gave the opportunity to continuously develop the reference design during the project. In this paper, the ‘final’ reference design is described as implemented in the Daimler truck, which took part in the verification, validation and demonstration of multi-brand platooning together with six other truck manufacturers.

So far, hydrogen compatibility of polymer materials was investigated, focusing on gaining insight in general degradation mechanisms. However, this is not yet sufficient for purposes of safety and performance requirements, for example, in the automotive industry and especially, the corresponding advanced material development, which needs meaningful and comprehensive data for prospective long time intervals. Therefore, herein, the applicability of forced aging regimes and the suitability of different analytical techniques for clarification of the underlying mechanisms are focused on. The study analyzes the behavior of polyether ether ketone, (thermoplastic) polyurethane and fluoroelastomer materials after prolonged aging in hydrogen atmosphere under pressure as it occurs in hydrogen fuel‐cell vehicles. Material changes are investigated by electron microscopy, thermogravimetry, differential scanning calorimetry, and gas sorption measurements. Especially thermogravimetry turns out to indicate subtle material changes that are important indications for material choice.

<div class="section abstract"><div class="htmlview paragraph">The European Commission plans to introduce a (solid) particle number (PN) emission limit for type approval and in-service conformity (ISC) by the end of 2018 (Euro VI d) using PEMS (Portable Emission Measurement System) tests on heavy duty vehicles on the road. Performance, measurement accuracy and sensitivity of several on-board particle counters for heavy duty applications have not been tested yet in parallel on a chassis dyno with Euro VI vehicle (N3-class, 12.8 l). The PN PEMS examined were CPC (Condensation Particle Counter) and DC (Diffusion Charger) based. Evaluation was conducted at different ambient temperatures from −7 °C to 35 °C while running different test cycles: WHVC (World Harmonized Vehicle Cycle), steady state engine operation, active regeneration and ISC-tests. A particle number system following the current heavy duty regulation requirement and recommendations of the Particle Measurement Program (PMP) served as reference (PMP_TP). CPC based instruments showed stable and accurate measurement behavior with good correlations to PMP_TP (CPC#1: s = 1.08, R<sup>2</sup> = 0.99; CPC#2: s = 0.74, R<sup>2</sup> = 0.97), DC based instruments showed good correlation at high concentrations (>1x10<sup>11</sup> p/kWh) (DC#1: s = 1.15, R<sup>2</sup> = 0.96; DC#2: s = 1.05, R<sup>2</sup> = 0.91). No particular influence was seen from regeneration, high exhaust gas temperature, high exhaust flow, ambient temperature. T<sub>amb</sub> = −7 °C affected one CPC instrument. Probably urea particles were counted by one DC based instrument because their size could be close to DC instruments cut-off size. The main conclusion of this study is that PN-PEMS are robust and testing is feasible for heavy duty applications.</div></div>

Functional near-infrared spectroscopy (fNIRS) provides reliable results for determining cognitive load based on averaged cortical blood flow during multiple repetitions of short cognitive tasks. At the same time, it remains unclear how to use this technique for assessing cognitive load during prolonged single-trial activity. In this study, we used a computer-based emergency simulation game for inducing different levels of cognitive load. We propose a novel approach to measure cognitive load using specific time slots, determined based on simulation log-data interpreted in light of Barrouillet's time-based resource-sharing model. To validate this approach, we compared cortical activity in dorsolateral prefrontal cortex (DLPFC) and left inferior frontal gyrus (IFG) regions measured at four specific time slots during a simulation. We found significant associations between cognitive load and neuronal activity within the DLPFC depending on the chosen time slot, whereas no such dependencies were found for the IFG. These results illustrate how knowledge of task structure could be used advantageously for the identification of cognitive load. Although requiring further investigation in terms of reliability and generalizability, the presented approach can be considered promising evidence that fNIRS might be suitable for more general reliable assessments of cognitive load during prolonged single-trial activities and for real-time adaptations in simulation-based learning environments.

Machine learning approaches for manufacturing usually offer recommendation lists, e.g., to support humans in fault diagnosis. For instance, if a product does not pass the final check after the assembly, a recommendation list may contain likely faulty product components to be replaced. Thereby, the list ranks these components using their probabilities. However, these probabilities often differ marginally, while economic impacts, e.g., the costs for replacing components, differ significantly. We address this issue by proposing an approach that incorporates costs to re-rank an initial list. Our evaluation shows that this approach reduces fault-related costs when using recommendation lists to support human labor.