Research On Vehicle Speed Prediction Model On Urban Road Ramp And Curved Road

With the further acceleration of urbanization and the increasing degree of urban motorization,urban roads in China have also maintained a stable development trend.In the case of the rapid development of urban roads,the traffic safety problems of urban roads are becoming more and more serious,especially in the slopes and curved road sections where the road alignment is not coordinated with the running speed of the vehicles.Therefore,in order to improve the road shape of the ramps and corners,the actual measured speed of the road will be used to predict the speed of small passenger cars on urban road ramps and curved road sections.This provides a theoretical basis for the linear optimization design of urban road ramps and curved road sections,and provides theoretical support for the speed limit of urban road ramps and curved road sections.Firstly,the slopes with different slopes and the curves of different corners in Chongqing are selected as experimental sections,and the speed is collected by the passenger cars equipped with GPS equipment and cameras,and the passenger cars of urban road ramps and curved road sections are obtained.The measured speed of the car.The collected vehicle speed is processed to obtain the speed of the passenger car in the urban road ramp and the curved road section of Chongqing under the free flow state and not affected by the traffic lights and the weather.Secondly,the collected speed of the passenger car is processed,including smoothing the data with five-point three-time smoothing method,correcting the missing data according to the speed of the data points before and after the missing data points,and screening the unqualified data.Combined with the measured speed of the measured vehicle,the influence of the slope of the up and down slope and the length of the ramp on the running speed of the small passenger car on the ramp is analyzed,as well as the influence of the radius of the curve,the length of the curve and the speed of the inflection on the running speed of the small passenger car on the curve.Then,after analyzing the characteristics of the speed change on the ramp,the slope of the upgrade,the slope of the downgrade and the length of the ramp which have a great influence on the speed of the urban road slope section are selected as the influencing variables,and the prediction models of the speed of the downhill section and the running speed of the downhill section of the minibus based on the regression analysis are established respectively,and the rationality of the model is analyzed.Finally,after analyzing the characteristics of vehicle speed change in curved roads,the radius,length and speed of curve which have a great impact on the speed of urban road curve section are selected,and the prediction model of car speed in urban road curve section is established by using GA-BP neural network algorithm.The reliability of the model is analyzed by combining the measured speed of actual road.The results show that the forecasting results of the models established in this paper are accurate,and can be used to predict the speed of urban road ramps and curved sections.These two models can be used in the optimization design of road alignment and speed limit of urban road ramps and curves.