The launch of both battery electric vehicles (BEVs) and autonomous vehicles (AVs) on the global market has triggered ongoing radical changes in the automotive sector. On the one hand, the new characteristics of the BEV powertrain compared to the combustion type have resulted in new central parameters, such as vehicle range, which then become an important selling point. On the other hand, electric components are as yet not optimized and the sensors needed for autonomous driving are still expensive, which introduces changes to the vehicle cost structure. This transformation is not limited to the vehicle itself but also extends to its mobility and the necessary infrastructure. The former is shaped by new user behaviors and scenarios. The latter is impacted by the BEV powertrain, which requires a charging and energy supply infrastructure. To enable manufacturers and researchers to develop and optimize BEVs and AVs, it is necessary to first identify the relevant parameters and costs. To this end, we have conducted an extensive literature review. The result is a complete overview of the relevant parameters and costs, divided into the categories of vehicle, infrastructure, mobility, and energy.

Battery research depends upon up-to-date information on the cell characteristics found in current electric vehicles, which is exacerbated by the deployment of novel formats and architectures. This necessitates open access to cell characterization data. Therefore, this study examines the architecture and performance of first-generation Tesla 4680 cells in detail, both by electrical characterization and thermal investigations at cell-level and by disassembling one cell down to the material level including a three-electrode analysis. The cell teardown reveals the complex cell architecture with electrode disks of hexagonal symmetry as well as an electrode winding consisting of a double-sided and homogeneously coated cathode and anode, two separators and no mandrel. A solvent-free anode fabrication and coating process can be derived. Energy-dispersive X-ray spectroscopy as well as differential voltage, incremental capacity and three-electrode analysis confirm a NMC811 cathode and a pure graphite anode without silicon. On cell-level, energy densities of 622.4 Wh/L and 232.5 Wh/kg were determined while characteristic state-of-charge dependencies regarding resistance and impedance behavior are revealed using hybrid pulse power characterization and electrochemical impedance spectroscopy. A comparatively high surface temperature of ∼70 °C is observed when charging at 2C without active cooling. All measurement data of this characterization study are provided as open source.

A significant disadvantage of battery electric vehicles compared to vehicles with internal combustion engines is their sharply decreased driving range at low temperatures. Two factors are primarily responsible for this decreased range. On the one hand, the energy demand of cabin heating needs to be supplied by the vehicle’s battery since less waste heat is available from the powertrain, which could be used to cover heating demands. On the other hand, a limited capability to recuperate at low temperatures serves to protect the battery from accelerated aging, which ultimately leads to less energy regeneration. This paper analyzes the impact of both factors separately on a battery electric vehicle’s driving range. Additionally, this paper provides technical requirements for the implementation of an electrothermal recuperation system. Such a system has the potential to reduce the impact of both abovementioned factors on driving range by enabling the direct usage of regeneratable energy for heating when battery charging is limited under cold conditions. The presented analysis is based on BMW i3 and Tesla Model 3 datasets, which combined cover more than 125 trips in and around Munich at different ambient conditions. The results show that the range can decrease by up to 31.9% due to heating and by up to 21.7% due to limited recuperation, which gives a combined maximum range decrease of approximately 50% under cold conditions. Additionally, it was found that a heater with a short reaction time in the lower millisecond range and a power capability of 20 kW would be sufficient for an electrothermal recuperation system to enable the utilization of unused regenerative braking potentials directly for heating.

In order to improve the driving range and reduce the cost of battery electric vehicles (BEV) through a higher efficiency, this paper proposes to adopt multilevel converters using low voltage Si MOSFETs in the electric powertrains. A multilevel Si MOSFET inverter, a conventional IGBT inverter and a SiC MOSFET inverter are modelled and compared using a reference vehicle over various driving cycles. The costs of the three solutions are also compared. It is shown that the multilevel Si MOSFET inverter has a rather high efficiency and realizes the lowest cost among the three solutions even when the worst case of cost is considered. Sensitivity analysis also shows that the multilevel Si MOSFET inverter is suitable for a wide range of vehicle concepts in addition to the reference vehicle. Moreover, the multilevel topology also features lower electromagnetic interference (EMI) and provides modularity. Therefore, Si MOSFET-based multilevel inverters are proved in this paper to be an appropriate option to improve the efficiency and reduce the cost of electric powertrains.

Automotive traffic scenes are complex due to the variety of possible scenarios, objects, and weather conditions that need to be handled. In contrast to more constrained environments, such as automated underground trains, automotive perception systems cannot be tailored to a narrow field of specific tasks but must handle an ever-changing environment with unforeseen events. As currently no single sensor is able to reliably perceive all relevant activity in the surroundings, sensor data fusion is applied to perceive as much information as possible. Data fusion of different sensors and sensor modalities on a low abstraction level enables the compensation of sensor weaknesses and misdetections among the sensors before the information-rich sensor data are compressed and thereby information is lost after a sensor-individual object detection. This paper develops a low-level sensor fusion network for 3D object detection, which fuses lidar, camera, and radar data. The fusion network is trained and evaluated on the nuScenes data set. On the test set, fusion of radar data increases the resulting AP (Average Precision) detection score by about 5.1% in comparison to the baseline lidar network. The radar sensor fusion proves especially beneficial in inclement conditions such as rain and night scenes. Fusing additional camera data contributes positively only in conjunction with the radar fusion, which shows that interdependencies of the sensors are important for the detection result. Additionally, the paper proposes a novel loss to handle the discontinuity of a simple yaw representation for object detection. Our updated loss increases the detection and orientation estimation performance for all sensor input configurations. The code for this research has been made available on GitHub.

There is rapidly growing interest in autonomous electric vehicles due to their potential in improving safety, accessibility, and environmental outcomes. However, their market penetration rate is dependent on costs. Use of autonomous electric vehicles for shared-use mobility may improve their cost competitiveness. So far, most of the research has focused on the cost impact of autonomy on taxis and ridesourcing services. Singapore is planning for island-wide deployment of autonomous vehicles for both scheduled and on-demand services as part of their transit system in the year 2030. TUMCREATE developed an autonomous electric vehicle concept, a microtransit vehicle with 30-passenger capacity, which can complement the existing bus transit system. This study aims to determine the cost of autonomous electric microtransit vehicles and compare them to those of buses. A total cost of ownership (TCO) approach was used to compare the lifecycle costs. It was shown that although the acquisition costs of autonomous electric vehicles are higher than those of their conventional counterparts, they can reduce the TCO per passenger-km up to 75% and 60% compared to their conventional counterparts and buses, respectively.