Road Network Regional Traffic States Analysis Based On Floating Car Data

In recent years,due to the acceleration of the urbanization process and the improvement of residents’ living standards,the number of cars has increased rapidly,and the problem of urban traffic congestion has become increasingly serious.The economic losses and traffic accidents caused by traffic congestion have many negative impacts on urban residents and society.Therefore,how to alleviate traffic congestion has become one of the hot research directions.The emergence of intelligent traffic control systems can effectively alleviate traffic congestion and greatly improve the quality of urban traffic services.Mining urban road network state information and partitioning the road network,to reduce the complexity of the road network has become an important research direction.These studies can discover the internal rules of urban traffic and provide a strong theoretical basis for traffic managers to formulate traffic congestion mitigation measures.This article uses floating vehicle GPS data to study and analyze the traffic state of the road network area,and its purpose is to accurately grasp the traffic state changes in the sub-region of road network.At the same time,this article gives an effective solution to the problems of data loss and noise caused by various reasons of floating vehicle GPS acquisition equipment.The main work of the thesis is as follows:(1)The method of determining the floating car sample capacity was used to study the feasibility of using the existing floating car GPS data in Xi’an to analyze the traffic status of the road network area.At the same time,the data preprocessing of the original GPS data containing errors and errors was performed.(2)To solve the problem of missing data and noise in the road network parameters calculation,the paper proposes Gaussian weighted KNN algorithm of traffic data filling,and solves the problem of missing data effectively;Then,a wavelet threshold noise reduction method is used to perform noise reduction on the completed traffic data,so as to ensure that the traffic data after noise reduction is relatively stable and the overall state trend of traffic road conditions are well reflected.(3)Aiming at the problems of instability and parameter sensitivity of traditional traffic road network division methods,the paper proposes an AP algorithm-based city traffic road network division method,and uses this method for traffic state recognition of urban road network and road network homogeneity regional division.By mining and analyzing the traffic state of the road network at different time scales,it is found that this method can effectively identify and predict the current congestion state and future change trend of the traffic network;by mining and analyzing the traffic time series data,the method can be successfully used to identify the homogenous areas,and find the bottleneck sections or sensitive nodes in road network.