Functional Safety Concept Generation within the Process of Preliminary Design of Automated Driving Functions at the Example of an Unmanned Protective Vehicle

Graubohm, Robert; Stolte, Torben; Bagschik, Gerrit; Steimle, Markus; Maurer, Markus · 2019 · Crossref

DOI: 10.1017/dsi.2019.293

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of structuring the early design phase for automated driving systems, specifically focusing on the systematic generation of functional safety concepts. The motivation stems from the increased complexity of driverless vehicle development, where safety considerations significantly influence design outcomes, particularly when human drivers are absent as a fallback. The authors aim to demonstrate how tasks and work products from the ISO 26262 safety life cycle can be mapped to a reference process during preliminary design, using the unmanned protective vehicle developed in the aFAS research project as a case study. The study employs a structured approach to functional safety concept generation within the concept phase of development. This process involves iterative loops connecting item definition, hazard analysis and risk assessment (HARA), and functional safety concept generation. The authors utilize the Goal Structuring Notation (GSN) to document the safety concept, providing a graphical representation that links safety goals, mitigation strategies, and safety requirements. This method improves traceability and readability compared to traditional tabular documentation. The case study involves an SAE Level 4 automated vehicle designed for low-speed operation on highway hard shoulders, following a leading vehicle. The analysis incorporates inputs such as operational design domains, system boundaries, and functional system architectures, while deriving safety goals and Automotive Safety Integrity Levels (ASIL) through HARA. The results present a comprehensive set of functional safety requirements and an excerpt of the safety argument structure for the aFAS vehicle. The authors identified 17 safety goals, with ASIL classifications ranging from QM to D, addressing hazards such as unintended steering actuation, mode changes, and obstacle detection. For instance, Safety Goal 3 (preventing steering actuation beyond specification) was classified as ASIL D and broken down into specific requirements involving steering angle limits and brake interventions. The GSN notation effectively illustrated how abstract safety goals are decomposed into concrete requirements assigned to functional components like the steering system, brake system, and human-machine interfaces. The study highlights that safety requirements are derived based on architectural assumptions and mitigation strategies, ensuring they allow for implementation. The significance of this work lies in providing a systematic process structure for determining safety requirements in automated driving functions, bridging the gap between functional concepts and technical design. By demonstrating the use of GSN for safety concept notation, the paper offers a method to create preliminary safety arguments that can be extended during development. The findings emphasize the importance of iterative refinement and the integration of behavioral safety aspects with functional safety mechanisms. This approach supports the efficient development of complex automotive systems by ensuring consistent and traceable safety conceptualization, contributing to the broader field of automated vehicle safety engineering.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• driverless ads