Prototype autonomous driving safety isolation on Arm Neoverse
Introduction
Why functional safety matters in software systems
Understand functional safety risks
Apply ISO 26262 and ASIL levels
Implement safety-critical isolation using safety island architecture
Functional safety for automotive software development
How to use Data Distribution Service (DDS)
Deploy OpenAD Kit across multiple cloud instances
Run OpenAD Kit across distributed ROS 2 instances
Next Steps
Prototype autonomous driving safety isolation on Arm Neoverse
Introduction
Why functional safety matters in software systems
Understand functional safety risks
Apply ISO 26262 and ASIL levels
Implement safety-critical isolation using safety island architecture
Functional safety for automotive software development
How to use Data Distribution Service (DDS)
Deploy OpenAD Kit across multiple cloud instances
Run OpenAD Kit across distributed ROS 2 instances
Next Steps
Review frequently asked questions
These FAQs were drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.
Place time‑critical, safety‑relevant control logic (for example, braking or steering) on the safety island, and keep non‑critical features (such as infotainment) on the general ECU. The goal is strong isolation, determinism, and minimized coupling for safety‑critical paths.
Check that safety‑critical components run independently from non‑critical services and communicate only through defined publish‑subscribe interfaces. Ensure data exchanged is minimal and purpose‑specific so that safety logic is not impacted by unrelated functions.
Higher ASIL targets require more rigorous processes and evidence across the V‑model. For example, ASIL‑D changes go through full impact analysis and regression testing to prevent introducing new risks.
Both approaches support prototyping: containers model software isolation on one system, while multiple instances model stronger physical separation. Choose the option that best matches the isolation assumptions you want to evaluate.
Maintain clear safety requirements, rationale for the safety‑island split, defined DDS/ROS 2 interfaces, and mapped tests to requirements. Record versioned changes, impact analyses, and verification results aligned to the V‑model stages.