Road intersection steering ratio prediction method based on LSTM neural network

Abstract

The invention discloses a road intersection steering ratio prediction method based on an LSTM neural network. The road intersection steering ratio prediction method comprises the steps: firstly carrying out the statistics of traffic flows in different flow directions of all entrance lanes of a road intersection, and carrying out the data preprocessing; then, calculating the steering ratios of allthe entrance lanes of the road intersection in all the flow directions, and splitting a training set and a test set; then, designing a prediction model based on an LSTM neural network, and selecting an appropriate activation function and an appropriate loss function; and finally, sequentially selecting the training set and the test set of different entrance lanes, training and testing the corresponding prediction model, and completing the establishment of the prediction model, thereby being applied to the prediction of the road intersection steering ratio. According to the road intersection steering ratio prediction method, the time sequence characteristic of the traffic flow can be fully utilized, and the prediction precision of the road intersection steering ratio is effectively improved, and meanwhile, the problems of gradient explosion and gradient dispersion are avoided through the reasonable neural network structure and the good activation function characteristic, and the applicability of the prediction method is improved.

Description

technical field [0001] The invention relates to the road intersection steering ratio prediction of intelligent traffic, and the road intersection steering ratio prediction is used for traffic signal control adjustment, traffic design and traffic planning. Background technique [0002] In the field of intelligent transportation, the prediction of traffic flow parameters is a very important research direction. If the relevant traffic flow parameters can be predicted more accurately, such as road section speed, traffic flow, travel time, road intersection steering ratio, etc., urban traffic managers can make traffic signal control adjustments in advance, or carry out reasonable traffic design and planning. , thereby alleviating urban traffic congestion and improving traffic efficiency. In the intelligent transportation system, the steering ratio of road intersections represents the ratio of the number of vehicles flowing in each direction of an entrance at a road intersection ...

AI Technical Summary

Problems solved by technology

This type of method has certain requirements on the smoothness of traffic flow, and is suitable for predicting the steering ratio of road intersections that have been smoothed for a long time. It is difficult for the road intersection prediction method of the model to make effective predictions and high-precision predictions

The second type of method is a prediction method based on Kalman filter, BP (Back Propagation) neural network, etc. Although this type of method has improved algorithm efficiency and prediction accuracy, it cannot effectively use the steering ratio of road intersections in the time dimension. Correlation and periodicity, there are large limitations

[0004] At present, the existing steering ratio prediction methods at road intersections have the following main problems: 1) Most methods use linear models to derive and predict the steering ratio of road intersections. Such methods require smooth traffic flow and cannot The actual traffic flow that adapts to nonlinear changes does not meet the actual traffic demand; 2) Other prediction methods based on Kalman filter and BP neural network, although the algorithm efficiency and prediction accuracy have been improved, but these methods completely ignore The correlation and periodicity of the steering ratio in time are limited, and its prediction accuracy is greatly limited

Images