Refine
Year of publication
Document Type
- Article (80)
- Master's Thesis (54)
- Bachelor Thesis (48)
- Diploma Thesis (30)
- Part of a Book (12)
- Book (5)
- Report (4)
- Other (2)
Institute
Language
- English (235) (remove)
This master's thesis focuses on developing a robust framework for co-simulating microscopic traffic scenarios and vehicle dynamics, leveraging the capabilities of SUMO and CARLA. The essence of this research lies in its meticulous analysis of existing simulation tools, leading to identifying the most effective co-simulation strategies and intercommunication methods. A comparative study elucidates the strengths and limitations of these methodologies, guiding the selection of an optimal approach.
Central to the methodology is a thorough understanding and application of SUMO and CARLA, enhanced by the strategic implementation of intercommunication methods. The detailed integration of these tools and the introduction of test automation significantly boost simulation efficacy. The framework's reliability and accuracy are rigorously validated through well-defined procedures and metrics, ensuring fidelity in simulation results.
Key to this research is the design of experimental scenarios that reflect real-world traffic conditions underpinned by carefully chosen simulation parameters. These scenarios are pivotal in validating the co-simulation framework's performance, supplemented by a performance cost analysis focusing on processing efficiency. The thesis culminates with a critical discussion of the findings, addressing the research objectives and posing solutions to identified challenges. Potential areas for future enhancements are explored, highlighting the framework's limitations.
In conclusion, this thesis contributes significantly to the field of traffic simulation for autonomous driving technologies. It presents a comprehensive co-simulation framework, offering researchers and engineers a sophisticated tool for advanced testing and validation. The insights and recommendations provided are instrumental for ongoing research and development in this rapidly evolving domain.
The book 'Reinforcement Learning: An Introduction' by Sutton and Barto is the standard text book for introductory courses to reinforcement learning. Next to concrete algorithms and extensive examples the book contains several fundamental results related to Markov decision processes (MDPs) and Bellman equations in Chapters 3 and 4. Unfortunately some proofs are missing, some theorems lack precise formulation, and for some results the line of arguments is quite garbled.
In this note we provide all missing proofs, give precise formulations of theorems and untangle the line of arguments. Further, we avoid using random variables and their expected values. Since we (like Sutton/Barto) restrict our attention to finite MDPs all expected values can be made explicit avoiding overloaded notation and murky conclusions.
This article bridges the gap between introductory literature like Sutton/Barto and research literature containing exact formulations and proofs of relevant results, but being less accessible to beginners due to higher generality and complexity.
This article reconstructs the path of the German economist Friedrich A. Lutz (1901–75) to American economics. The correspondence with his former teacher Walter Eucken, the founder of the Freiburg school, constitutes a crucial yet unexplored source for the article. Through Lutz's case, this article demonstrates the growing gulf between German and Anglo-Saxon economics during the late 1930s. In his native Germany, Lutz was trained in methodologically and institutionally focused economics, which differed fundamentally from the mathematical economics dominating Anglo-Saxon academia. He realized that an academic career in the United States would be impossible if he did not adapt to the new methods and if he did not abandon the methods of the German tradition. This gave rise to his internal Methodenstreit. After his emigration in 1938, he constantly experienced doubts and tensions because he was convinced that without considering institutions, mathematical economics could never explain the occurrence and essence of macroeconomic phenomena. Despite his stellar career at Princeton, it was only after his move to Zurich in 1953, where he taught history and theory of socioeconomics for the rest of his life, that Lutz could reconcile this internal Methodenstreit.
OpenStreetMap (OSM) is a large open database for geographic data created and maintained by volunteers. OSM's main data use is rendering an extremely detailed map of the world. Data quality is an important issue for applications like routing of pedestrians to public transport facilities. In this report we describe different schemes for mapping bus stops in OSM and we provide statistics on usage of those schemes, the good ones and the not so good ones.
Sustainable Integration of Digitalisation in Nursing Education - An International Scoping Review
(2024)
The book is the second of four volumes on data science and artificial intelligence. This second volume covers data visualization tool and techniques as well as fundamentals of supervised machine learning: linear regression, artificial neural networks, support-vector machines, decision trees, ensemble methods and more.
Model predictive control (MPC) is a promising approach to the lateral and longitudinal control of autonomous vehicles. However, the parameterization of the MPC with respect to high-level requirements such as passenger comfort, as well as lateral and longitudinal tracking, is challenging. Numerous tuning parameters and conflicting requirements need to be considered. In this paper, we formulate the MPC tuning task as a multi-objective optimization problem. Its solution is demanding for two reasons: First, MPC-parameterizations are evaluated in a computationally expensive simulation environment. As a result, the optimization algorithm needs to be as sample-efficient as possible. Second, for some poor parameterizations, the simulation cannot be completed; therefore, useful objective function values are not available (for instance, learning with crash constraints). In this work, we compare the sample efficiency of multi-objective particle swarm optimization (MOPSO), a genetic algorithm (NSGA-II), and multiple versions of Bayesian optimization (BO). We extend BO by introducing an adaptive batch size to limit the computational overhead. In addition, we devise a method to deal with crash constraints. The results show that BO works best for a small budget, NSGA-II is best for medium budgets, and none of the evaluated optimizers are superior to random search for large budgets. Both proposed BO extensions are, therefore, shown to be beneficial.
This paper embeds the early political economy of Friedrich August von Hayek in the intellectual milieu of German ordoliberalism. The urgency during the 1930s and 1940s to stabilize the disintegrating societal orders is identified as a crucial driver behind the parallelisms between Hayek and the ordoliberals. Their shared theoretical position is that in such moments, liberty can thrive sustainably only after a framework of general and stable rules has been established. Hayek’s proximity to ordoliberalism was most explicitly discernible in The Road to Serfdom and at the founding meeting of the Mont Pèlerin Society in 1947, culminating in the shared politico-economic vision of the competitive order. The contextual nature of Hayek’s ordoliberalism surfaced in the years after The Constitution of Liberty when his focus shifted, along with the postwar intellectual and institutional stabilization of the West: from how stable orders enable liberty, to how liberty enables the evolution of orders.
The labor market is confronted with social, environmental, and economic developments that affect working conditions and individual labor relations. Lately, the Covid-19 pandemic has demonstrated and reinforced the importance of inclusive growth and sustainable work relationships. In this chapter, the smart job factory, a metamodel that supports the creation of new forms of work by redefining roles in labor, is introduced. The smart job factory is based on social entrepreneurship principles to drive innovative, sustainable, and long-term solutions to social challenges. For practical application, the model can be translated into a software solution that supports employers in the assessment of current working conditions and job roles within their companies and helps to redefine work relationships and to create new jobs. Thus, the smart job factory supports the labor market transition by systematically and proactively shaping new forms of work based on the triple bottom line of sustainable development.
The traffic calming measures aim to slow down traffic speed, accident frequency, and reduction of through traffic caused mainly by motorized vehicles in residential areas. These measures are primarily addressed to the specific streets and lead the necessary or remaining traffic to drive in a restrained or considerate manner. While these measures are designed to prevent conflicts between pedestrians and motorists, they impose unstable traffic patterns and are sometimes unable to accommodate the increasing motor vehicle flows, as they are concentrated only on certain streets.
This paper investigates area-wide urban traffic calming techniques from existing projects. It explains how traffic regulatory and structural measures link together or separately to restrict the movement of motor vehicles traveling through neighborhood streets and divert them to main roads. The implemented sample measures were illustrated and described with photos of their current locations.
In this thesis, the district Äußere Neustadt was defined as a model area in Dresden to analyze and redesign the existing mobility plan in order to reduce the MIV flows with the main objectives of giving space for pedestrians, cyclists, and public transport. The district's street infrastructure is heavily impacted by the presence of parked vehicles, raising challenges to the safe movement of both bicyclists and oncoming motor vehicles. In the frame of the model project initiative - "Woche des guten Lebens", the volunteer team has designed a traffic experiment and carried out an online survey to assess citizens' opinions of the Äußere Neustadt. The analyses demonstrate and emphasize the necessity of political support and interactive communication with citizens regarding the area-wide radical sustainable mobility plan. Considering the positive feedback of the citizens, the new mobility plan was realized in this paper.
Area-specific traffic calming techniques such as diagonal or cross barriers, zone speed limits, offsets, one-way streets, etc. were analyzed from the implemented projects to determine the appropriate solutions for the specified streets. The district's existing traffic network and the new solutions offered were visualized using QGIS software.
The implemented new traffic plan will lead to more space for pedestrians and bicycle traffic and a reduction in traffic noise.
The objective of the study is to improve the functionality of the Virtual Reality Laboratory (VR Labs) of the Faculty of Automotive Engineering and enhance its applicability in Road Traffic Engineering education and training at the University of Applied Sciences Zwickau (WHZ). In this study, a comprehensive description of Virtual Reality (VR) and Augmented Reality (AR) as well as their wide application in education has been given.
An insight into the VR Labs of the Faculty of Automotive Engineering at WHZ is provided. The study built and proposed eight use cases of VR Labs in road traffic engineering training and education at WHZ. The limitations and barriers in deploying application cases are analyzed and suitable solutions are given. The adoption of virtual reality can be an easy transition based on BIM modeling software development, and 3D and 360 degrees scanning.
An extension of the VR Labs application for other areas of education at WHZ has been identified. In addition, technologies, areas of application, and future research directions are mentioned. Innovations such as 3D modeling and BIM-based workflows that are being encouraged through the development of content-created engines will provide students with opportunities for using VR/AR on road traffic and infrastructure projects in the future.
BIM (Building information modeling) is becoming a reliable method for the planning, design, implementation, and maintenance stages of construction projects due to its ability to improve the quality of project stages, reduce project time, and ensure costs. From 2000 onwards, the trend of using BIM method increased in the world. And in the last decade, many developed countries have introduced BIM as a reasonable and efficient method with the aim of optimizing project stages, and many employers have paid attention to training and persuading consulting and contracting companies.
In the meantime, the BIM method is constantly being updated. Many researchers are looking to modify existing sub-methods to help the project achieve all of the great BIM goals.
Although BIM has great goals such as those mentioned, in some cases, projects are not able to achieve all of them for various reasons.
This thesis has put its hypothesis on the existence of deficiencies in the content of the EIR (Employer's Information Requirements) document and its writing methods.
This document, which is a kind of project guide in BIM format, is compiled by the client and its task is to announce the needs of the client to the members involved in the project and to clarify the different stages of the project. In many construction projects, due to the complexity of the EIR, the employer's lack of attention to some information, or the failure to include the opinions and views of the project's stakeholders in the writing of this document, the EIR cannot fulfill its key role with its maximum potential performance in the project.
This thesis is written with the aim of finding a comprehensive solution to maximize the power of the BIM method in a project through correct and complete document writing.
Also, this thesis seeks to prove its claims by focusing on the ABS 38 project as a case study. In addition, a literature review on important and key issues in the field of BIM and EIR documents has been done. After that, a scoring system (Relative rating method) was used to find the best variant and compare it to prove the proposed hypothesis.