Research On The Automatic Generation Method And Its Application For Dynamic Topology Of Nanning Road Network Based On Real-Time Trajectory Data

Understanding the traffic network and its dynamic traffic flow structure characteristics is very important for scientific planning and management of urban traffic.The real-time traffic trajectory data contains the traffic flow information of urban roads.Nowadays,mining traffic network and its traffic flow dynamic structure characteristics through real-time trajectory data has become a research hotspot all around the world.Therefore,based on the actual requirements of traffic management in Nanning,this paper proposes an automatic generation algorithm of road network dynamic topology based on real-time traffic trajectory data and electronic map,which provides important basis for traffic network optimization,traffic guidance and other applications.It mainly completes the following work:Firstly,it investigates the research situation of traffic trajectory data processing and urban traffic network topology,and discusses the difficulty and emphasis of the automatic generation of traffic trajectory data and road network topology.And then,the research framework is constructed with utilizing weighted directed graph mode for traffic network modeling and weighted adjacency matrix to represent the model.In Second,it studies the real-time map matching algorithm of trajectory data,designs the data pretreatment method of real-time trajectory and GIS data and uses refined grid to reduce the search space of roads in order to achieve rapid positioning of the vehicle position data.Besides,with the mapping relations between multi-grid and roads as the grid index,it reduces the time complexity of map matching algorithm.Based on this works,it realizes the real-time map matching of trajectory data of more than 6700 floating Cars in Nanning.Then,it designs the automatic generation algorithm of urban road network topology.Based on the electronic map road information,with road intersection as topological node and the connected road between the nodes as the topological edge,it establishes a graph model named “node-link”;it introduces the concept of road group and generates weights for topological edge dynamics according to the traffic flow operation state of road group.Furthermore,it realizes the automatic generation of urban traffic network weighted dynamic directed topological graph and achievesoptimal path recommendation and its travel time estimation with the use of the traffic network topological graph.Finally,it verifies the model and algorithm and actually applies this method to Nanning traffic simulation platform.The experimental results show that the proposed method in this paper can achieve the real-time processing of traffic trajectory data in Nanning,generate urban backbone topology automatically and meet the accuracy and real-time requirements of traffic information service.