Application Of Vehicle Satellite Positioning Data In Urban Road Traffic Safety Evaluation

With the vigorous development of China’s economy and society and the orderly progress of the urbanization process,road transportation infrastructure is gradually improved,the number of motor vehicles continues to increase,and the role of the transportation industry in the national economy is increasingly prominent.However,urban road traffic safety has always restricted the development of urban road traffic.Identifying road safety levels and taking targeted improvement measures are of great significance for improving road safety.Traditional road traffic safety assessments are mostly based on traffic accident data.However,traffic accident data is usually not easy to obtain,especially the amount of new road data is limited.The thesis proposes a road traffic safety evaluation model based on vehicle satellite positioning data.Considering the differences in traffic characteristics of road sections and intersections,the road sections of urban roads were re-divided.Using Python language to achieve the vehicle satellite positioning data acquisition and preprocessing of a city’s online car rental,coordinate transformation of the vehicle satellite positioning data,and use OSM open block map to complete the map matching process.These data preprocessing and road network redefinition methods have laid the foundation for accurately realizing the visualization of vehicle satellite positioning data trajectory points in digital map sections.An abnormal driving behavior recognition framework based on satellite positioning data is proposed to recognize abnormal driving behaviors such as overspeed,rapid acceleration and sudden deceleration,and the influence of threshold changes on the recognition results of abnormal driving behavior is analyzed.The results show that the lower the threshold setting,the greater the number of abnormal driving behaviors that can be identified.The number of sudden acceleration abnormal driving behaviors is lower than that of overspeed and sudden deceleration abnormal driving behaviors,while the number of abnormal driving behaviors of overspeed and sudden decelerations is higher,which is consistent with previous studies on the analysis of operating characteristics of operating vehicles.Establish a road safety evaluation model based on vehicle satellite positioning data and neural network classification methods.Taking the number of abnormal driving behaviors of speeding,rapid acceleration,and rapid deceleration as model input variables,and taking the road safety level obtained by k-means clustering as the output of the model,the training set is used to learn and train the BP neural network model,and the trained BP neural network model is used as the classification model of the safety level of urban road sections.The research results show that the neural network classification model constructed using vehicle satellite positioning data and urban road section accident data has a certain predictive ability and can be used to classify the safety level of urban road sections,so that traffic management departments can find dangerous sections of urban roads.It can be applied to road traffic safety improvement practices.