Connected and autonomous vehicles (CAVs), as a critical component of the modern transportation system, have demonstrated significant potential in enhancing traffic efficiency, ensuring safety, and improving environmental quality, making them a research hotspot in the field of intelligent transportation. This paper conducts an in-depth review of the latest research achievement in vehicle platoon models, platoon control strategies, and data-driven system architectures. It also provides a comprehensive overview of the current status and development trends in data-driven platoon control for CAVs. Firstly, the paper summarizes the methods for modeling vehicle platoons and compares the similarities, differences, and advantages of mechanistic modeling and data-driven modeling. Secondly, it delves into the optimization control issues in the platoon system, with a focus on analyzing the differences between traditional control methods and data-driven control strategies, particularly highlighting the application potential of multi-objective optimization methods in intelligent and connected traffic environments. Next, it elaborates on the data sources, system architecture and implementation pathways of data-driven control strategies. Finally, the paper identifies the key challenges and bottlenecks of the development of platoon control systems and outlines future research directions and priorities, aiming to provide both theoretical guidance and practical insights for promoting the widespread application of CAVs in complex traffic environments.