基于高危场景的 ADAS 与自动驾驶预期功能安全 (SOTIF) 仿真验证平台
SelfDrivingSim 专注于自动化发掘自动驾驶系统在真实世界运行中的“未知不安全场景”(从 SOTIF 区域 3 判定缩减至区域 4 安全已知域)。基于 OpenSCENARIO 2 声明式约束求解内核,平台在数百万维参数化长尾空间中执行组合变异模糊测试,为整车厂提供符合 ISO 21448 标准的权威合规认证。
SOTIF Scenario Runner Zonal Architecture
Central Controller: SOTIF Scenario Runner — ISO 21448 Compliance Engine
Zonal Control Units & TSN Ring
- ASAM OpenSCENARIO 2.0 Interpreter
- Combinatorial Boundary Fuzzer
- Criticality & TTC Safety Metric Evaluator
- Zonal Control Software Under Test (SUT)
- Automated SOTIF Compliance Auditor
系统性攻克自动驾驶出行中致命的‘未知不安全’ (Unknown-Unsafe) 场景
与专注于传统硬件电气/电子随机失效的 ISO 26262 功能安全不同,ISO 21448 预期功能安全 (SOTIF) 专门解决系统在硬件完全无故障情况下,因环境复杂多变或算法认知感知局限性引发的严重危害事件。SelfDrivingSim 采用主动式边界搜索算法,自动化变异光照对比度、几何轨迹航向角及动态障碍物突发遮挡,全面圈定自动驾驶算法模型的安全失效包络线。
Core Engineering Areas
Technical Articles
Related Cluster Architecture Specifications
- AutomotiveSDV — Full SDV Stack & Reference Architecture
- A2HCard — Agent-to-Human (A2H) Interface Cards & In-Cabin Autonomous Interaction Manifests
- A2HumanCard — Standardized In-Vehicle AI Agent Capability Discovery, Trust & Authorization Cards
- AIDrivenSynBio — AI-Driven Synthetic Biology for Automotive In-Cabin Air Purification, Bio-Engineered Pathogen Scavengers & Molecular Biosensors