“With the increasing dependence of advanced intelligent driving on multimodal perception, prediction, and decision-making, real traffic data faces bottlenecks such as high collection costs, insufficient coverage, and difficulty in annotation in extreme weather, long tail scenes, and privacy sensitive environments, making it difficult to support system scale training and validation. How to efficiently generate traffic data with realism and controllability to improve the reliability of the system in extreme situations has become a key issue that urgently needs to be addressed. Based on this, this article provides a systematic review of traffic data intelligent generation technology for advanced intelligent driving, aiming to grasp the research progress and guide engineering practice. Firstly, we introduce the typical process of model system evaluation and define and analyze the core challenges currently faced, such as data scarcity, cross modal alignment, controllable conditions, scene consistency, and closed-loop verification; Subsequently, around the technical frameworks of diffusion models, generative adversarial networks, neural radiation fields/3D Gaussian splashes, world models, and multimodal large models, representative generation methods and their key applications and technical points in the three major application directions of intelligent cockpits, single vehicle intelligent driving, and multi vehicle collaborative perception based on vehicle road collaborative perception were systematically sorted out; Finally, a multi-level evaluation framework covering perception prediction control closed-loop measurement and sensor physical consistency was proposed, and several practical suggestions for building an engineering data engine that combines authenticity and diversity were discussed. The algorithms, datasets, and evaluation metrics mentioned in this article have been summarized in https：//github.com/fayewong666999/higher-level-smart-driving-data-generation。 This article aims to provide a systematic reference for the data system construction, evaluation standards, and subsequent technological evolution of advanced intelligent driving.”