FSP2: Mobilität
Refine
Year of publication
Document Type
- conference proceeding (article) (152)
- Article (99)
- conference proceeding (presentation) (23)
- Preprint (14)
- Part of a Book (13)
- conference proceeding (volume) (8)
- Book (3)
- Other (3)
- Doctoral Thesis (2)
- Report (2)
Language
- English (248)
- German (71)
- Multiple languages (3)
Publication reviewed
- begutachtet (271)
- nicht begutachtet (49)
Keywords
- automated driving (14)
- Glocke (12)
- advanced driver assistance systems (10)
- Simulation (9)
- human factor (8)
- workload (8)
- ACDC (7)
- driving experience (7)
- Automated Driving (6)
- lane keeping system (6)
Institute
- IFM - Institut für Fahrerassistenz und vernetzte Mobilität (107)
- Fakultät Elektrotechnik (86)
- Fakultät Maschinenbau (56)
- Fakultät Tourismus-Management (40)
- INIT – Institut für Nachhaltige und Innovative Tourismusentwicklung (40)
- Fakultät Informatik (28)
- ECC-ProBell - Europäisches Kompetenzzentrum für Glocken (26)
- IMS - Institut für Maschinelles Sehen (18)
- IFA – Institut für Innovative Fahrzeugantriebe (16)
- ILE – Institut für Leistungselektronik (15)
Der Beitrag zeigt die Modellierung, Simulation und den Test des Einflusses von Regen und Nebel auf Messungen mit LiDAR- und Radar-Sensoren. Zunächst wurden relevante Kriterien zur Modellierung hergeleitet und erläutert. Die Anwendung der Theorie der Mie-Streuung und Rayleigh-Streuung führte zu einer Entwicklung von LiDAR und Radar Sensor Simulationsmodellen. Diese Modelle wurden durch Tests in Regen-Testananlagen validiert. Hierbei liefern Key Performance Parameters (KPIs) quantitative Ergebnisse wie Signal Dämpfung, Signal-Rauschverhältnis (SNR), Detektionsrate, Fehlerrate, Entfernungsfehler.
Light detection and ranging (LiDAR) sensors are increasingly applied to automated driving vehicles. Microelectromechanical systems are an established technology for making LiDAR sensors cost-effective and mechanically robust for automotive applications. These sensors scan their environment using a pulsed laser to record a point cloud. The scanning process leads in the point cloud to a distortion of objects with a relative velocity to the sensor. The consecutive generation and processing of points offers the opportunity to enrich the measured object data from the LiDAR sensors with velocity information by extracting information with the help of machine learning, without the need for object tracking. Turning it into a socalled 4D-LiDAR. This allows object detection, object tracking, and sensor data fusion based on LiDAR sensor data to be optimized. Moreover, this affects all overlying levels of autonomous driving functions or advanced driver assistance systems. However, since such sensor-specific effects are rarely available in public datasets and the velocities of target objects are not included as ground truth in these datasets, it makes sense to enrich the limited real-world data with synthetic data. Therefore, this paper discusses how such datasets can be created and combined to efficiently predict velocities on realworld data using the authors' novel method dubbed VeloPoints.
At the current development stage, the lower airspace above urban areas is only used to a very limited extent. Recent developments in the drone industry are making this area more accessible with the possibility to set up so-called U-Spaces, where this area is controlled for manned and unmanned aerial participants. The leading use case for drone applications is currently seen in the medical sector. The relevance of the use of medical drones in urban areas and in which conditions the technology will bring specific advantages is still unclear.
Autonomous driving and traffic flow simulation requires a realistic and accurate representation of the environment. Therefore, this research focuses on the semantic segmentation of aerial images for simulation purposes. Initially, a dataset was created based on true orthophotos from 2019 and Kempten’s street cadaster, with true orthophotos being fully rectified aerial images. The chosen classes were oriented towards the subsequent conversion and usage in simulation. The proposed labeling workflow used cadaster data and demonstrated significant time efficiency compared to state-of-the-art datasets. Subsequently, a neural network was implemented that was trained and tested on the dataset. In addition, the network was also trained only on the lane markings to compare the network’s performance. Both cases demonstrated excellent segmentation results. The generalizability was then tested on true orthophotos from 2021. The results indicated a solid generalizability, but still needs to be improved. Finally, the aerial information was converted into a 3D environment, that can be used in simulations. Our results confirm the usage of aerial imagery and street cadaster data as a basis for the simulations.
The development of Automated Driving Systems (ADS) has the potential to revolutionize the transportation industry, but it also presents significant safety challenges. One of the key challenges is ensuring that the ADS is safe in the event of Foreseeable Misuse (FM) by the human driver. To address this challenge, a case study on simulation-based testing to mitigate FM by the driver using the driving simulator is presented. FM by the human driver refers to potential driving scenarios where the driver misinterprets the intended functionality of ADS, leading to hazardous behavior. Safety of the Intended Functionality (SOTIF) focuses on ensuring the absence of unreasonable risk resulting from hazardous behaviors related to functional insufficiencies caused by FM and performance limitations of sensors and machine learning-based algorithms for ADS. The simulation-based application of SOTIF to mitigate FM in ADS entails determining potential misuse scenarios, conducting simulation-based testing, and evaluating the effectiveness of measures dedicated to preventing or mitigating FM. The major contribution includes defining (i) test requirements for performing simulation-based testing of a potential misuse scenario, (ii) evaluation criteria in accordance with SOTIF requirements for implementing measures dedicated to preventing or mitigating FM, and (iii) approach to evaluate the effectiveness of the measures dedicated to preventing or mitigating FM. In conclusion, an exemplary case study incorporating driver-vehicle interface and driver interactions with ADS forming the basis for understanding the factors and causes contributing to FM is investigated. Furthermore, the test procedure for evaluating the effectiveness of the measures dedicated to preventing or mitigating FM by the driver is developed in this work.
Tourism is an important economic driver for numerous regions, at- tracting more than one billion visitors annually. While economically significant, excessive numbers of visitors lead to local overcrowding, which negatively im- pacts visitors’ experience and safety, and causes environmental harm. This paper proposes a practical approach to empowering destination management organiza- tions (DMOs) to manage tourist flows. We advocate for a rule-based approach that models visitor occupancy based on easily understandable influence factors like weather and date. As a central component, an ontology-guided knowledge graph ensures compatibility with diverse touristic data models and allows seam- less integration into existing infrastructures. By digitizing DMOs’ experiential knowledge, we facilitate the implementation of lean and cost-effective visitor guidance. We demonstrate our approach by implementing two applications for two different use cases. The results of our qualitative evaluation reveal the com- pelling potential for rule-based occupancy modeling approaches serving as a baseline for future visitor management systems.
Many cities in Europe and around the world are concerned with reducing their CO2-emissions. One step on this agenda is the introduction of electric buses to replace combustion engines. The electrification of urban buses requires an accurate prediction of the energy demand. In this pa per, an energy model and the underlying calibration process is presented. This approach leverages substantial tracking data from 10 electric buses operated in Göttingen, Germany. It was shown that, with the use of additional information from the directly measured tracking data, like auxiliary power, charging power and vehicle weight, it is possible to precisely calibrate models based on physical equations with regard to generally poorly identifiable parameters like rolling friction coefficient or efficiency of the electric machine. With a multilayered approach for simulating the energy demand, it is possible to validate the results on the mechanical layer and on the electrical layer separately. This enables a far better parametrization and elimination of uncertainties from the different parameters. Furthermore, we compare the results to other publications for sections with 1 km, 100 km and 230 km, respectively. The relative errors between the simulated and measured electrical power consumption are below 0.3%, 3% and 6.5%, respectively. Hence, the yielded model is appropriate for electric urban bus network planning applications. And the found parameters should be a good starting point for other energy prediction models. To further enable comparability with other approaches the dataset used for calibration is made publicly available.
In the early phase of new vehicle system developments, it is crucial to fully define and optimize working system and functional architectures. Architecture definition and validation in turn requires a quick and accurate evaluation of a system‟s overall performance. Modeling and simulating a complete vehicle system, however, is complex and in many cases was either technically not achievable or simply has been omitted within the development process. It is the utmost challenge in system modeling and simulation to realistically reflect interaction of various electrical, mechanical, thermal, and software elements as attributed to individual system modules and their relations. State-of-the-art tools meanwhile bear this capability. In this paper we present an approach how they may effectively and efficiently be incorporated into a car system development process. To accomplish this target, we „virtualize‟ all system entities while defining and reflecting all relevant system aspects. Our proposed development flow allows simulating, evaluating, and validating complete vehicle systems and their behavior. The proposed flow will sustainably change car system development processes.
Safety of the Intended Functionality (SOTIF) addresses sensor performance limitations and deep learning-based object detection insufficiencies to ensure the intended functionality of Automated Driving Systems (ADS). This paper presents a methodology examining the adaptability and performance evaluation of the 3D object detection methods on a LiDAR point cloud dataset generated by simulating a SOTIF-related Use Case. The major contributions of this paper include defining and modeling a SOTIF-related Use Case with 21 diverse weather conditions and generating a LiDAR point cloud dataset suitable for application of 3D object detection methods. The dataset consists of 547 frames, encompassing clear, cloudy, rainy weather conditions, corresponding to different times of the day, including noon, sunset, and night. Employing MMDetection3D and OpenPCDET toolkits, the performance of State-of-the-Art (SOTA) 3D object detection methods is evaluated and compared by testing the pre-trained Deep Lea rning (DL) models on the generated dataset using Average Precision (AP) and Recall metrics.