Analysis Of The Spatio-Temporal Characteristics Of Residents 'Travel Based On Beijing's Taxi Trajectory Data

As a highly concentrated area of residents' living activities, is city's architectural pattern, urban function planning and transportation planning appropriate? They are a direct influence on the sustainable development of the city, the efficiency of people's work and the happiness index of life. At the same time, in this complex system, the individual's activities (ie, the relationship between the individual and the space, interactive relationship between people and nature) have become the focus of a large number of scholars.Due to the influence of the inherent regional functional distribution of the city, residents' daily travel activities will show a certain spatio-temporal mode (cycle mode,gathering mode, etc.). In order to further understand the distribution of spatio-temporal characteristics of urban residents, this research takes Beijing as an example to study the spatio-temporal characteristics of residents' activities on the weekdays and weekends by using the taxi trajectory data.The main work includes: (1) By pre-processing the taxi track data,we extracted the taxi OD data; (2) The statistical results of the residents'travels on the weekdays and weekends are analyzed. (3) The taxi data of one day is divided into six periods(7:00-10:00?10:00-13:00?13:00-16:00?16:00-19:00?19:00-22:00?22:00-00:59), and Kernel Density is performed for the OD point of each period. Based on this result we have a preliminary analysis of the residents' travel. (4) Considering the advantages and disadvantages of each clustering algorithm and the characteristics of the taxi track data along the network distribution, a hybrid clustering method based on grid density is proposed. The algorithm is used to cluster the OD point of taxi at different periods.Then, in conjunction with POI data, the density distribution of POI data in the clustering results is counted. Based on the above results, the relationship between the activity area of the residents in different periods and the functional type of the area is analyzed.The final results show that there is a significant difference between weekdays and weekends. The weekday presents a clear morning and evening peak travel characteristic, while the weekend has no prominent peak travel. Through the analysis of the clustering results, it is concluded that the weekday travel is mainly commute, and the distribution of travel density shows the characteristics of concentric circles. It is focus on the residential areas and the work areas, and leisure area has increased the heat after 16:00.The travel of the resident experienced such a process(spatial relative dispersion- spatial aggregation- spatial relative dispersion).In addition, the research result is of great significance to analyze the residents' travel patterns and urban planning. Finally, we summarize the article, and point out the shortcomings of the study, and look forward to the future improvements in this research.