Architecture design of automated driving system based on RFLP method

To address the problems of the lack of a global perspective, difficulties in knowledge transfer and reuse, and the inability of existing modeling methodologies that rely on a single modeling language to meet the requirement that multi-level collaboration and multi-disciplinary modeling in automated driving system design, a requirement-function-logic-physics-based automated driving system modeling methodology is proposed. A multi-level modeling process covering requirement analysis, functional analysis, logical architecture, and physical architecture design is built. A hierarchical mapping relationship is established to facilitate the layer-by-layer transfer and traceability of requirements to system architecture, supporting subsequent model-based system verification. Modeling verification is conducted on the automated driving systems, and the results showed that the proposed methodology effectively improved the normalization, consistency, and traceability of modeling, providing methodological support for the design of complex automated driving systems.