Study Of Urban Trip Mode Discrimination Under Mobile Network Environment

Transportation, which serves as a link connecting the city and transports peopleand goods, is essential for the development of a city. It is the main driving force ofcity progress. In this day and age, economic and technology has made a big leap andthe development of the city is facing a new problem in the new situation. Intelligenttraffic systems, which act as new modes of transportation, should be implementedimperatively. The21st century will be the century of intelligent transportation.The key to develop intelligent transportation is to get rich traffic information.Traffic information, as the basis of intelligent transportation, serves the daily urbantraffic management and transport services. It draws great consideration with the needof urban development. Trip mode is serves as an important traffic informationdatabase and it has become a hotspot in the field of transportation planning.Discriminating urban trip mode effectively not only provides a reliable data sourcesand evaluation, but also has important implications for solving problems such asurban congestion, spatial structure optimization, the energy consumption and so on.During the urban trip mode discrimination method research, the most criticalissue is the data acquisition of residents,travel. With the development of computer,communications, information technology, smart mobile phones become more popular.Traditional travel data collection methods such as questionnaires, home visits, recalllogs cannot meet the demand of traffic management. People pay attention to get traveldata by using new technologies and new methods under the latest technologies andsocial background. This thesis proposes a travel information collection method basedon mobile phone positioning in the mobile network environment and using Java todesign a data acquisition system for intelligent terminals. The data acquisition systemis simple to install and use. The type of data has collected extensive which containsboth static and dynamic information. While the data collection performs, the systempreprocesses the data first and pre-determined the travel path. Compared to traditionaldata collection methods, the new travel information collection method has manyadvantages such as less investment, high-precision, all-weather real-time acquisition, powerful data preprocessing capabilities and so on. People use this method for themassive real-time extraction information and then conduct trip mode discriminationstudy.When studying the trip mode discrimination, this research is not restricted tothe traditional fuzzy judgment. It also strives to achieve a finer division oftransportation on the travel path. Combining data collection methods and data transferpoint, the thesis puts forward the apriority sequential clustering segmentationalgorithm for segmentation of the travel path on the basis of trip modepre-classification. The combination of theory and exploratory methods improves theaccuracy of extracting different transportation modes greatly. Since then, according tothe results of the travel path segment, the article deeply analyzes the characteristics ofurban trip mode and discriminates the trip mode by using the machine learningalgorithms decision tree without reliance on secondary data (such as bus stations, buslines, subway lines, road network, etc.) as well as knowledge of relevant experiencesituations. Besides, the discriminant analysis method is verified by the test data andreaches a very high accuracy.Finally, the paper establishes a trip discriminant model based on the theoreticalapproach. Adding a trip discriminant module to independently developed i4Peopleplatform, it will lay a solid foundation to put research results into practice.