Bus arrival time forecasting method based on mass historical GPS (global position system) trajectory data

Abstract

The invention discloses a bus arrival time forecasting method based on mass historical GPS (global position system) trajectory data. In the bus arrival time forecasting method, a bus route is divided into multiple travel sections and the average travel time of the travel sections is calculated according to the mass GPS trajectory data; according to the ratio of the distance between the current position of a bus and the nearest bus station to the total distance of the current travel section, and on the basis of the average travel time of the current travel section, the arrival time of the bus to the nearest bus station can be forecasted. On the basis of the mass GPS trajectory data, stable and referable data of the different travel sections can be acquired, and the forecasting precision is 87.6%.

Description

technical field [0001] The invention relates to a bus arrival time prediction method, in particular to a bus arrival time prediction method based on massive historical GPS track data. Background technique [0002] The bus arrival time refers to the travel time from the current position of the bus running along the bus line to the target stop. With the development of social economy and the rapid increase of the number of motor vehicles, urban traffic is facing great challenges. Prioritizing the development of public transportation and realizing intelligent public transportation is not only an important part of the intelligent transportation system, but also an effective way to solve urban traffic problems. The research and development of high-performance bus arrival time prediction technology is of great significance to realize the intelligentization of the bus system on the basis of management and service information, and to improve the management and service level of urban...

AI Technical Summary

Problems solved by technology

The neural network model can fit historical data well, but its accuracy mainly depends on the length of training time, so the real-time prediction is poor

[0006] At the same time, most of the existing bus arrival time prediction models are based on massive traffic records, focusing on mathematically establishing solutions and parameter equations to predict the arrival time of the target object and other related data. The calculation process is complex , it is difficult to further improve the prediction accuracy

Images