| Time | Title | Speaker |
|---|---|---|
| 13:30 | Opening Remarks | |
| 13:35 | Legal Requirements for Programming Autonomous Driving Functions - Ensuring Compliance With Traffic Regulations | Hans Steege (CARIAD / University of Stuttgart) |
| 14:15 | Democratising Safety: Safety Engineering with Safety Pool Studio | Guild Bruce (WMG Warwick) |
| 14:55 | Traceable Behavior for Automated Vehicle - An MBSE-Inspired Approach | Marcus Nolte (TU Braunschweig) |
| 15:35 | Coffee Break | |
| 16:00 | Risk Prediction Methodology Based on Near-Miss Incident Database | Pongsathorn Raksincharoensak (Tokyo University of Agriculture and Technology) |
| 16:40 | SafeADArchitect: End-to-End Architecture for Safe and Risk-Aware Automated Driving | Fabian Oboril (Intel Labs Germany) |
| 17:20 | Closing Remarks |
The presentation provides an overview of the legal requirements for automated and autonomous driving functions from the German Road Traffic Act (StVG). Following the amendments to the StVG in 2017 and 2021, Sections 1a ff. of the StVG and Sections 1d ff. of the StVG now contain provisions for highly and fully automated and autonomous driving. Especially the behaviour of the driving functions is relevant for operating vehicles. The speaker will focus on compliance with and programming of traffic regulations.
As Automated Driving Systems (ADSs) become more prevalent, effectively communicating safety to the general public is more crucial than ever. In this talk, we will explore the relationship between communicating safety and engineering safety, by showcasing how Safety Pool Studio democratises safety and links public scenario creation to the concrete safety testing of autonomous vehicles. This session will also highlight the work of PAVE UK in educating diverse audiences about ADS safety more generally.
Specifying how automated vehicles should behave in public traffic is still a challenge, particularly when it comes to traceability requirements that are related to scenario-based safety analyses which are required in the scope of SOTIF. This talks presents an approach that combines domain knowledge with established concepts from model based systems engineering. It presents an architecture framework whose views and viewpoints enable the consequent tracing of resulting vehicle behavior to design decisions and requirements.
The presentation will describe the construction of near-miss incident database collected from taxi vehicles in Japan. The database is used for analysis of human error, accident mechanism, the development of advanced driver assistance systems and automated driving. Nowaday, digital twin technology and artificial intelligence are also applied to reconstruct critical driving scenarios for evaluating the performance of ADAS/AD.
One roadblock for mass deployment of automated vehicles (AVs) in crowded and unstructured environments such as urban traffic is the high degree of uncertainty that significantly reduces the vehicle's usability using today's safety solutions. To tackle this challenge, this talk presents SafeADArchitect, an end-to-end architecture for automated vehicles that ensures usability and safety at the same time. It covers major influencing factors to driving safety from the driving platform, sensor limitations or behavior uncertainties through occlusions, inefficiencies of AI-based perception or planning approaches via adequate safety modules and by mapping possible errors to associated driving risks. Hence, by knowing the overall acceptable risk, reasonable driving decisions can be taken that balance usability and safety.