Automotive Architecture Topologies: Analysis for Safety-Critical Autonomous Vehicle Applications

Safety-critical systems such as Advanced Driving Assistance Systems and Autonomous Vehicles require redundancy to satisfy their safety requirements be classified fail-operational. Introducing in a system with high data rates processing also has great impact on architectural design decisions. The current self-driving vehicle prototypes do not use standardized architecture but base existing vehicles the available components. In this work, we provide novel analysis framework that allows us qualitatively quantitatively evaluate an in-vehicle topology compare it others. With framework, different variants of two common topologies: domain zone-based architectures. Each is evaluated terms total cost, failure probability, communication cable length, load distribution, functional distribution. We introduce selected parts using our automated process provided safety-oriented enables ISO26262 Automotive Safety Integrity Level decomposition technique. After every step, re-evaluated. advantages disadvantages are guide designer towards choice correct architecture, focus introduction redundancy.