Research on Freeway Travel Speed Prediction Based on Commercial Vehicle Data

With the rapid development of China's economy and the continuous improvement of people's living standards, the public's travel demand is increasing. However, rapid traffic demand growth is bound to bring problems such as traffic congestion and traffic accidents. Accurate travel speed prediction can provide a basis for traveler route selection and traffic guidance. This study uses the Beidou positioning navigation system data to take the Shanghai-Kunming freeway section in Jiangxi Province as a test section to study the travel speed prediction. The contents include:(1) Summarizing the main methods of short-term traffic flow prediction. (2) Design the preprocessing process for Beidou system data. (3) A travel speed prediction model based on support vector regression is established. (4) The SVR model is used to conduct empirical prediction research on different traffic flow states.