The Research On Traffic State Identification And Travel Time Reliability Of Urban Road Based On License Plate Data

With the expansion of city scale and rapid development of urban economy,road traffic load and traffic congestion phenomenon is increasingly serious,and the contradiction between supply and demand of traffic is increasingly prominent.Due to these problems,the intelligent transportation system has been proposed to improve transportation efficiency,to improve traffic congestion situation and to ensure security running of vehicles.The intelligent transportation system can provide more effective means of transportation management,and make the traffic provide better services for people.In recent years,many Chinese cities began to use the traffic information collection based on license plate recognition technology to assist in road traffic management.The license plate recognition technology can get one kind of real-time monitoring date of vehicles driving on the urban road.These data are contiguous and large amounts,which have some features such as temporal and spatial correlations and so on.According to the lack of researches on intelligent transportation system based on license plate recognition data at present,the high-cost license plate recognition system can only achieve some simple functions such as traffic surveillance,traffic flow detection and so on,which leads it to be a lower price-to-performance ratio.So,how to make full use of the license plate recognition data for intelligent transportation related research is a current hot spot.Firstly,this paper studies the characteristics of current some traffic information collection technologies,license plate recognition system and data collected.With the analysis of possible fault data and missing data,this paper gives the corresponding methods to identify and modify these data,and concludes the travel time estimationafter data processing.Secondly,a method of traffic state identification based on license plate recognition data is proposed,which realizes the traffic state identification based on fuzzy theory.The method proves to be actually effective,which is tested using the license plate recognition data of a road in Qingdao.Thirdly,this paper analyzes the travel time reliability based on the license plate recognition data,and finds that urban road travel time conforms to logarithmic normal distribution.The nature of log-normal distribution function and the index of travel time reliability are analyzed.This paper selects the buffer index as the travel time reliability index,and determine the general process through travel time reliability analysis based on license plate recognition data.A simulation analysis based on the license plate recognition data of a road in Qingdao is made,and verifies the effectiveness of this algorithm.