Automated Compliance Testing Case Generation for Intelligent Connected Vehicles: A Multi-Layer Knowledge-Augmented Framework

Intelligent connected vehicle (ICV) testing faces significant challenges due to growing complexity and the need to comply with multiple regulatory standards. Traditional manual template-based and single-standard approaches suffer from limited scenario coverage, compliance risks, and inefficiency. This paper introduces the Multi-Standard Collaborative Compliance Framework (MSCCF), which integrates cross-domain standards through dynamic regulatory knowledge graphs, dual-engine formal verification, compliance-guided generative models, and multidimensional test generation. Our framework combines retrieval-augmented generation (RAG) with rule-constrained reasoning to create compliance-guaranteed test cases, while enabling comprehensive scenario simulation through hierarchical task decomposition (HTD). Experimental validation demonstrates superior performance in compliance accuracy (F1-score: 0.94) and scenario coverage (89.7%) compared to baseline methods, establishing a scalable approach for multi-standard ICV testing across seven critical dimensions: road geometry, infrastructure status, dynamic objects, environmental conditions, digital communication, occupant states, and vehicle health.