VANETs have been envisioned to be useful in road safety and many commercial applications. The vehicle node in VANET has many moving status, such as normal driving, waiting for traffic lights, which has great impact on many applications in VANETs. In the paper, the Data fusion method based on Rough Sets Theory is presented, which can integrate data from different sources such as GPS data and digital map data, and provides a more accurate Classification for vehicle moving status. The model architecture and an example are also presented.