With the increasing complexity and demand for intelligence in transportation system, the development of autonomous transportation systems has become a key approach to addressing current transportation challenges. This paper presents the design of a digital twin simulation platform for autonomous transportation, which consists of a physical space, a digital space, and a data center. By implementing data collection and physical-virtual mapping, the platform can reflect the dynamic state of the physical space in real-time, forming a closed-loop for autonomous control. Additionally, the platform can leverage simulation software to model various complex scenarios and predict motion states, providing a safe testing environment that enhances the platform’s integration, intelligence, and adaptability. Finally, this paper illustrates the platform’s application in adaptive cruise control, path planning, and video detection using a 2 m × 2 m micro-scale sandbox, offering a theoretical and practical foundation for the research and development of autonomous transportation systems.