Short-Term Traffic Flow Prediction Based on ANFIS

Abstract

Accurate short-term traffic flow prediction has become a critical problem in intelligent transportation systems (ITS). In the paper, a kind of adaptive prediction method for short-term traffic flow based on ANFIS (adaptive-network-based fuzzy interference system) model was presented. ANFIS is a fuzzy interference tool implemented in the framework of adaptive network. It combines the comprehensibility of fuzzy rules and the adaptability and self-learning algorithms of neural networks. The traffic flow prediction model with 104 changeable parameters will be established through the training process, the goal of which is reduce the prediction errors between real predicting output the ANFIS model and the desired output. The result of simulation research demonstrates that this method has the advantage of high precision and good adaptability. This scheme is novel and advanced in the domain of the road traffic flow prediction. The application of the scheme will remarkably improve the response efficiency and precision degree of the road traffic inducement and control system in our country.