The Scale Estimation Of Shared Automomous Vehicles Based On Urban Vehicle Demand

With the rapid development of science and technology,shared autonomous vehicles(SAV)have gradually replaced manual driving cars,making it possible for shared vehicles to serve urban trip demand.SAV can alleviate urban traffic congestion,parking difficulties,and other traffic problems by reducing the total number of vehicles on the road.And it is the inevitable trend of urban traffic development in the future.Therefore,it is necessary to research the scale of SAVs to meet the trip demands of all vehicles in the city.This thesis starts from license plate recognition data to explore a more efficient algorithm for extracting vehicle trip demand and,on this basis,calculates the initial scale of SAV.And a case of Jinhua city in Zhejiang Province is researched.The main research content of this thesis is summarized as follows:(1)The algorithm for extracting urban trip demand is proposed.Firstly,based on the license plate identification data of urban intelligent detectors,in order to identify the multitrip trajectory of the vehicle,a homomorphic linear clustering algorithm based on the motion state of the trajectory element is proposed by introducing the velocity correlation between adjacent trajectory elements and combining with the average velocity about trajectory elements,duration and direction change angle of the trajectory.Secondly,combined with the spatial distribution characteristics of urban point-of-interest(POI),the origin and destination of trip are matched to the traffic analysis zone(TAZ).Finally,by comparing with the traditional time interval method,the validity and accuracy of the algorithm are illustrated.(2)The research of scale for car-sharing under the condition of autonomous driving.Firstly,based on shared travel,by constructing the vehicle-sharing network,the problem of calculating the minimum vehicle scale is transformed into solving minimum path coverage of vehicle-sharing network,and the algorithm of Hopcroft Karp(HK)is introduced to solve the problem.Secondly,based on the shared travel,vehicle-sharing network under the combined travel mode is constructed to estimate the scale of SAV when shared autonomous vehicles and rail transit coexist,and further analyze the impact of rail transit on the scale of SAV.(3)The empirical research on Jinhua,Zhejiang Province.Firstly,based on the algorithm of extracting urban trip demand,more than 400,000 vehicle trips about one day are extracted.Secondly,after replacing the motor vehicles with SAVs,the minimum scale of SAV needed to meet all vehicle trip demands is calculated to be 62,284,which is 74.7% less than the total number of existing vehicles.Finally,based on the combined travel mode of SAV and rail transit,the minimum scale of SAV is calculated to be 49,345,which is 80%less than the total number of existing vehicles.The experimental results show that under the combined travel mode,the number of SAV is further optimized by 21%.The research results of this thesis are helpful to optimize the scale of SAVs and have practical guiding significance for determining urban traffic management and planning under the mode of Maa S in the future.