Road Traffic Risk Safety Prediction Based on BP Neural Network

In the field of smart transportation, the prediction of road traffic accident is an essential part. It analyzes the collected traffic information and predict the occurrence of traffic accidents in order to achieve effectively early warning and reduce ultimately traffic accidents. Generally, the traffic information includes speed, weather, light and road conditions. These data are usually not quantified. Thus, in this paper, those data will be quantified before they are able to be used as input parameters for the prediction. Based on data of historical traffic accidents and their statistical results in the UK, an accident severity model is established by regression analysis. A BP neural network optimized by genetic algorithm is utilized for predicting the accident severity under the current conditions. Via comparing both results, a road traffic safety risk prediction model is established. In order to improve the prediction accuracy of the BP neural network, three parameters for testing are presented and analyzed. Finally, the results show that the prediction of road traffic risk with optimized BP neural network is reasonable. It provides an effective safe driving for route planning.