Traffic demand estimation method based on GPS navigation data

Abstract

The invention discloses a traffic demand estimation method based on GPS navigation data, a big data technology and an artificial intelligence algorithm are increasingly mature, the traffic informationacquisition cost is lower than that at any time in the past, and resident travel characteristics, space-time distribution characteristics of traffic demands and the like can be more intuitively and clearly understood. According to macroscopic data characteristics presented by big data, a hierarchical clustering algorithm is utilized to divide hot spot areas required by resident travel. Scientificand reasonable clustering division can analyze the travel demand of current residents, and an effective basis is provided for relieving urban traffic congestion and maximizing the utilization of traffic resources. The accurate estimation of the traffic demand can better understand the characteristics of urban traffic behaviors, provides a theoretical basis for urban traffic planning, effectivelyreduces the travel cost of residents, and improves the dispersion capability of urban emergencies.

Description

technical field [0001] The invention relates to the field of transportation, in particular to a traffic demand estimation method based on GPS navigation data. Background technique [0002] With the acceleration of urbanization development in my country, urban traffic congestion and traffic pollution are becoming more and more serious, and traffic accidents frequently occur. Timely and accurate acquisition of traffic big data, visual analysis of data, and construction of traffic data processing models are the prerequisites for building intelligent transportation. GPS navigation data is one of the important sources of traffic big data. The mobile phone map APP records the user's GPS track, including time and latitude and longitude information. For example, Baidu Maps has 300 million monthly active users, 200 million kilometers of daily navigation services, covering 375 large and medium cities and 99% of national roads. Therefore, by analyzing the data of GPS navigation, a m...

AI Technical Summary

Problems solved by technology

The taxi travel demand is estimated by the grid division method, the spatio-temporal distribution characteristics of the taxi travel demand in different functional areas are analyzed, and the Kalman filter model, the neural network model and the combined model are used to predict the taxi travel demand in the software park. The defect of the method is that the determination of the size of the traffic grid unit depends on the size of the research area and the data characteristics of the research problem. There is no unified standard and basis, and the randomness is relatively large.

Images