Embedded Executive LIVE
summary
Discussion focuses on integrating AI inference engines with deterministic safety on a single chip, showing that ISO 8800 adds a layer atop ISO 26262 rather than replaces it. Key claims: memory and interconnect contention cause safety functions to stall, requiring architectural fixes at silicon level; safety monitors must become probabilistic monitoring instead of periodic checks; safety cases need realistic probabilistic evidence. Also flags related security standards such as ISO 21314.
AI-written summary · about
related talks
20mThe Challenges in Supporting Multiple Safety Levels in the Same System
57mWriting Safety Critical Automotive Software for High Perf AI Hardware
17mDeveloping a RISC-V Automotive Safety Island
41mTurning the Ignition on Safety: Zephyr RTOS in Automotive Compliance
16mAutomotive Solution Empowered by RISC-V Based Security and Functional Safety Module
11mKeynote: Shaping the Future of Automotive Computing with RISC-V
21mRISC-V and Functional Safety
09mDemo: Enabling Automotive Safety with Andes RISC-V IP
