Research on short-term traffic flow prediction model based on RNN-LSTM

Abstract

With the rapid development of economy and the increasing number of cars, it is the first task to effectively predict the traffic flow of road traffic to alleviate road congestion and reduce traffic accidents. Aiming at the uncertainty and nonlinearity of traffic flow data, the traffic flow prediction model based on long-term and short-term memory is designed and compared with the traditional BP model and RNN model. The simulation results show that the LSTM model has a higher prediction accuracy and better prediction effect on traffic flow.