Survey of LLMs and AI Agents in V2X: Simulation, Analysis & Architectures

In this paper, a comprehensive survey of the integration of large language models (LLMs) and AI agents in (Vehicle-to-Everything) V2X systems is presented. Their application in simulation, analysis, and architectural design is examined, highlighting the enhancement of realistic scenario generation, adaptive data processing, and real-time decision-making. Interdisciplinary perspectives, including cloud computing and data engineering, are explored to demonstrate scalable simulation frameworks and efficient analytics. Comparative analyses of existing platforms illustrate significant improvements in safety, efficiency, and reliability in modern transportation networks. Emerging trends and future research directions, such as scalable deployment, privacy preservation, and standardization, are discussed. The findings underscore the transformative potential of integrating AI technologies in V2X infrastructures, paving the way for smarter, safer, and more resilient transportation systems. These advances offer promising avenues for further innovation in mobility.