Travel Time Prediction Model for Urban Road Network based on Multi-source Data

Abstract

In view of the deficiencies of single data source for travel time prediction, multi-source data are used to improve the precision of travel time. Floating car and fixed detector are commonly used in traffic data collection, and they have certain complementarities in data types and accuracy. Therefore, the real-time traffic data of these two detectors are used as input parameters of prediction model, and Kalman filtering theory is used to establish travel time prediction model of urban road network. Finally, the model is simulated by Vissim 4.3 and the simulation results show that the average absolute relative error of travel time based on multi-source data is 5.18%, and it is increased by13.4% comparing with fixed detector data and increased by 7.2% comparing with floating car data.