Visual Analysis For Intelligent Transportation

In order to improve the comprehensive efficiency of the transportation system,China has further coordinated all kinds of transportation.By continuously improving the operation mode of transportation such as roads,railways and civil aviation,China has provided people with more convenient,fast,green and economic travel choices.Although the development trend of transportation integration can better meet people’s diversified travel needs,it also aggravates the complexity of comprehensive transportation network and brings troubles to people’s travel planning.At present,most of the research work on path planning focus on a certain kind of transportation network(such as urban public transportation network and air-rail combined transportation network),but there is a lack of research based on comprehensive transportation network,which cannot provide path planning schemes between cross-city locations.Therefore,based on comprehensive traffic data,this thesis studies the method of cross-city path planning,and investigates relevant visualization technologies in order to optimize the presentation of path planning data.The main work includes the following four aspects:1.According to the characteristics of the data source website,a comprehensive traffic dataset is made.The data is crawled through Requests library provided by Python3 and Selenium tools.In order to improve the efficiency of data collection,a scheme for avoiding crawler mechanism,a scheme for handling interruption of data collection and a scheme for multi-thread concurrent data collection are developed.2.A point-to-point multipath planning algorithm is proposed to solve the cross-city path planning problem in the comprehensive transportation network.Firstly,the P-Space model is used to construct the urban traffic network.Then,an improved Depth First Search is used to search the accessibility paths between cities.Finally,the Top-K path planning scheme is searched by combining the travel time constraint and travel cost constraint.Experimental results show that the proposed algorithm can provide a path planning scheme in line with the actual situation and meet the real-time requirements of users for the planning system.3.An edge bundling algorithm based on homologous control points is proposed,aiming at the visual clutter problem of visual layout of of multi-path planning scheme.Based on the topology structure information of the graph,the algorithm can calculate homologous control points and selects edge control points by using the shortest path algorithm.Then the degree of edge aggregation is optimized with the thinking of gradation.Finally,the edges are smoothed through Bezier curves and colored according to the direction of the edges.Experimental results show that the proposed algorithm balances the bundling degree of the whole and local paths in the visual layout of the path planning scheme,and can effectively reduce the visual clutter.4.A visual analysis platform for point-to-point multipath selection is designed and implemented,aiming at the presentation and selection of multipath planning scheme.The visualization platform mainly includes three functional modules: information input module,path planning scheme visualization module and path related information visualization module.At the same time,the corresponding interactive operation is designed for each function module,which improves the efficiency of the user to choose the path scheme.