Am Montag, den 24. März 2025, um 16:15 Uhr hält
Lars Klitzke
Universität Oldenburg
im Rahmen seiner beabsichtigten Dissertation einen Vortrag mit dem Titel
From Real-World Traffic Data to Scenarios in the Context of Automated Vehicles
Der Vortrag findet online statt
studconf.uol.de/rooms/gx0-zsk-6nf-ja3/join
Abstract:
The large-scale introduction of automated vehicles on public roads is an ambitious and challenging goal. This technology aims to significantly contribute to increased traffic safety and comfort, while also serving as the foundation for further innovative mobility concepts. However, one of the most significant challenges in introducing such systems is ensuring their compliance with local regulatory requirements, since it is essential that these systems function correctly and safely within complex real-world environments without a human fall-back option.
Due to the complexity of the systems and their environment, the driving function or the automated vehicle is systematically tested in specific scenarios. However, the availability of an extensive dataset with diverse scenarios is a critical prerequisite, which can be defined based on different data sources. The collection of traffic data in the real world, using either vehicles or (quasi-) stationary infrastructure, represents a particularly valuable source, as it realistically reflects the behaviour of human road users and also includes atypical behaviour or even critical conflicts between participants. Specifically, infrastructure-based traffic data collection allows a detailed description of scenarios and, due to continuous data collection, the identification of rare phenomena. However, this continuous stream of traffic data must be systematically processed to create a comprehensive collection of scenarios.
This thesis addresses this issue and presents a methodology for representing traffic data collected in the real world based on scenarios and their systematic identification from real-world traffic data. A hierarchical data model is used for this purpose, which semantically describes traffic data at four levels of abstraction. Various approaches to defining and identifying phenomena at those abstraction levels using real-traffic data are presented. In particular, the level of primitives is motivated and a methodology is proposed that allows traffic data to be represented in terms of primitives and, based on them, the derivation of manoeuvres. Based on this, a methodology is presented for the systematic definition and extraction of scenarios using ontologies. The different methods are integrated into a modular platform that enables the continuous identification and analysis of scenarios in real-world traffic data.
The procedures and methods presented in this work are individually evaluated using real-world problems. Overall, the results show that the proposed methodology enables the systematic identification and representation of scenarios from real-world traffic data, thereby contributing to the compilation of a comprehensive knowledge base of scenarios. Furthermore, the results highlight the versatility of the proposed methodologies, demonstrating their application to a range of research questions. The exemplary use of ontologies may also serve as a foundation for future work, particularly within the domain of artificial intelligence research.
Betreuer: Prof. Dr. Frank Köster