A Case Study of Continuous Assurance Argument for Level 4 Automated Driving

Automated driving has a wide-ranging impact on our lives, making it crucial to form a consensus among various stakeholders for the dependability, including safety and availability. Moreover, a continuous assurance argument is essential to cope with changes and uncertainties. In this paper, based on existing development and operational documents and logs in an automated driving system company, we create assurance cases for level 4 automated driving safety in a Japanese city using Goal Structuring Notation (GSN). The GSN diagram includes results from safety analyses such as STAMP/STPA and test results. Additionally, as a step toward continuous assurance argument, we have prototyped a system for monitoring operational parameters such as speed of oncoming vehicles, ensuring the values remain within the assumed ranges. The GSN diagram with other documents such as safety analysis documents and operational monitoring data will be available via a web-based GSN tool.