A Traffic Flow Prediction Method Based on Road Clustering and Bidirectional LSTM

Abstract

The invention discloses a traffic flow forecasting method based on road clustering and double-layer bidirectional LSTM, including: 1) When there is a missing value in the training data, the missing value is filled in the way of surrounding average for the missing value, Improve prediction accuracy; 2) Propose a method of correlation clustering of roads based on historical traffic data, divide roads into several groups, and use time information and spatial information at the same time in the data preprocessing stage to improve prediction accuracy; 3) Design a A two-layer bidirectional LSTM deep neural network model, which improves the prediction accuracy of the model; 4) proposes a method for batch training and testing the network model to speed up the training and testing speed of the neural network model; 5) proposes a Multi-model fusion method to improve prediction accuracy. The invention simultaneously improves the prediction speed and accuracy of the deep neural network in traffic flow prediction.

Description

technical field [0001] The present invention relates to technologies such as deep learning, traffic flow prediction, neural network design, and cluster training of neural network models, especially a deep neural network model based on road clustering and double-layer LSTM (Long Short-Term Memory, long-term short-term memory) The traffic flow forecasting method has guiding significance for improving the accuracy and speed of traffic flow forecasting. Background technique [0002] With the advent of the era of big data, real-time traffic network data in big cities is gradually increasing. As one of the most critical applications in the current popular unmanned driving technology, artificial intelligence traffic prediction can make reasonable predictions on traffic conditions after considering the relationship between time and space. , to help vehicles choose the most suitable route, especially in the case of urban congestion, the route selection has more practical significance...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the current situation that the accuracy and speed of traffic flow prediction by neural network are difficult to improve and some deficiencies in existing designs, a traffic flow prediction method based on road clustering and double-layer bidirectional LSTM deep neural network model is proposed. Using the information hidden in the data and realizing the multiplexing of important information, the correlation clustering of roads improves the accuracy of traffic flow prediction, and avoids the communication and synchronization between nodes of the neural network model during cluster training, which improves the traffic flow prediction accuracy. The speed of cluster training

Images