Identifying Critical Roads And Urban Hotspots Based On Taxi GPS Trajectories

With the rapid increase in the number of urban motor vehicles,urban traffic problems are becoming more and more serious.As an important public transport vehicle,taxi has the characteristics that the running route and time are completely determined by passengers and the trajectory data are real-time.Therefore,taxi trajectory data can well reflect the spatial and temporal patterns of residents' travel rule and urban traffic status.Using taxi GPS trajectory data to study urban traffic problems has become a research hotspot at home and abroad.Based on this,this paper will use the taxi trajectory data to study the travel rules of urban residents,critical roads critical crossroads and urban hotspots.The main contents are as follows:(1)Statistical analysis the travel behavior of Lanzhou residents from the aspects of travel length,travel time,total daily travel volume and travel volume of each period.The results show that urban residents prefer to travel short distances and the number of passengers on Sunday is significantly less than Monday to Saturday.Based on the congestion ratio,the urban traffic status is analyzed.The results show that there are obvious differences between the traffic status on weekdays and weekend days.(2)Based on the topological structure and traffic flow characteristics of urban road network,critical crossroads identification model of urban traffic network based on directed weighted complex network is established and the DWNodeRank algorithm is used to identify critical crossroads.The experimental results show that the algorithm can reasonably identify critical crossroads in the road network.(3)The new dual algorithm of directed weighted network is used to establish the spatial and temporal model of urban traffic network.According to the road network topology and the influence degree of traffic state between adjacent road sections,the mixed influence-based identication algorithm of critical roads is presented.The results show that,on the one hand,critical roads are basically stable,on the other hand,critical roads are also changed due to the different functional structure of the urban area.The above results have important guiding significance for the design of tidal lanes in urban traffic network.Finally,the correlation coefficient is used to validate the effectiveness of the recognition algorithm based on mixed influence.(4)Combining taxi trajectory data with grid partitioning method,a directed weighted network model is constructed and the DWNodeRank algorithm is used to identify urban hotspots.The experimental results show that this recognition algorithm not only can correctly identify urban hotspots,but also simplify the time complexity of hot spots identification.