一种基于动态递归神经网络的交通流量实时预测方法

Traffic Flow Real-time Prediction Using a Dynamic Recurrent Neural Network Model

智能交通系统是目前世界上公认的解决城市交通拥堵问题的最佳措施 ,实时、准确的交通流量预测是智能交通系统实现的关键技术之一。提出了一种基于改进型 Elman神经网络的交通流量实时预测方法 ,由于预测模型中采用的递归神经网络具有动态记忆能力 ,因而可在网络规模较小的情况下实现对交通流量的快速、准确预测 。

Intelligent transportation system (ITS) is recognized as one of the best ways to solve the problem of traffic jam and traffic safety in cities. Accurate real-time prediction of traffic flow is the key technology in ITS. A traffic flow prediction model using modified Elman neural network (NN) is put forward in this paper. Because of its dynamic memory, the proposed recurrent NN model can predict traffic flow fast and correctly in the condition of smaller network size or fewer neurons. The presented prediction approach is proved to be useful and effective with simulation results.