Study On Freeway Traffic Volume Prediction Model Based On Floating Car Data

In the past 30 years,economic development of China has been speeded up remarkably.At the same time,traffic congestion has become increasingly serious.Traffic congestion not only affects the quality of travel,but also easily causes traffic accidents and energy waste,which is not conducive to the sustainable development of society.To some extent,traffic congestion will be avoided by grasping the real-time traffic status and taking timely traffic guidance measures.In order to collect traffic information,road detectors have been widely installed on expressways,such as coil detectors,infrared detectors,video detectors,etc.But the above traditional detection methods have some disadvantages,such as high maintenance cost and poor real-time performance.With the rapid development of GPS technology,floating car has become a new data acquisition method widely used in the transportation field.This paper established a traffic volume prediction model based on GPS floating car data,which makes it possible to obtain real?time road traffic information.Firstly,this paper introduced the development of GPS technology and several traffic volume prediction models.By analyzing the advantages and disadvantages of previous traffic volume prediction models,we established the prediction model based on GPS data and traffic flow theory.Secondly,this paper reviewed three important traffic flow parameters.We establish a simple road network model based on the location of ramp and detector.In order to reduce data deviation,K-Means clustering algorithm is used.We use the least squares method to fit the traffic flow parameters.And then the velocity-density relationship of each section is determined.Then,the calculation process of traffic volume prediction model is introduced.We established a non-linear programming function based on traffic flow theory.The optimal value obtained by this function is the number of vehicles in two adjacent floating cars.The cumulative traffic volume can be obtained by calculating the average density.This part is the core of this paper.Finally,we established expressway simulation model and generate the floating car data by VISSIM simulation platform.We input the floating car data into prediction model to validate the applicability of the model.By comparing with the actual traffic volume,it is found that the predicted results of the model are in line with the actual traffic conditions.We also found the estimated value of this model is closer to the real situation by comparing with the simple prediction model,and the prediction effect is relatively good.