Model of Delay Prediction for Signalized Intersection Based on GPS Data

Ya WANG, Chang-jun JIANG, Hong-lei REN

Abstract


To facilitate travel, many domestic bus transits have been fit with station prompt means. With this function, the bus arrival time can be forecast in real time based on the GPS data. However, the current bus arrival time prediction model features high assumption, not well adapted to urban roadway traffic congestion and delay at signalized intersection, thus compromising the accuracy of the prediction. A real time delay estimation for signalized intersection is proposed in this paper, combining the physical model derivation and data fitting, integrating the waiting time at signalized intersections into the existing prediction model. It is shown in the experiments that compared with previous models, the accuracy of our model increases by 29.1%. The mass GPS data is processed via the Hadoop platform, the performance is improved by 3600%~4000% compared with the serial program.

Keywords


Model of Delay Prediction for Signalized Intersection Based on GPS Data

Publication Date


2016-11-17 00:00:00


DOI
10.12783/dtetr/amita2016/3689

Full Text:

PDF

Refbacks

  • There are currently no refbacks.