A Short-Term Traffic Flow Prediction Method Based on Kernel Extreme Learning Machine

Real-time and accurate traffic flow prediction plays an important role in ITS (Intelligent Transport System). Extreme learning machine (ELM) has proven to be an efficient and effective learning paradigm for a wide field. With the method of kernel function instead of the hidden layer, Kernel-ELM overcame the problem of variation caused by randomly assigned weights. In order to improve the accuracy of traffic flow prediction, this paper introduces a kernel extreme learning machine(KELM)-based forecasting method. We have implemented KELM to forecast real-time traffic flow in the data of Nanning in South China. The results indicate that the KELM method can generate a more accurate prediction and provide better performance on extremely fast learning speed compared with other state-of-art algorithms.

I. Introduction

Prediction of short-term traffic flow is considered as a challenging problem in ITS(Intelligent Transport System). Accurately forecasting traffic flow is quite significant for effective traffic management system. Real-time and accurate traffic flow prediction has a wide range of applications, including traffic control, traffic induction, and vehicle routing, etc. The problem induced by the randomness, nonlinearity and complexity of traffic flow has compelled us to search for more reliable techniques to forecast traffic flow.