< Back to Projects

Using sparse gps data for route choice estimation

Project Number: 200

Project Leader: Anders Karlström

Start Date: 2011-09-01

End Date: 0000-00-00

Frågeställning GPS data are becoming increasingly available to be used in transportation research. On a massive scale, it is not always feasible to collect GPS data from individual cars with high frequency (each second, say). For instance, band width considerations impose restrictions such that the position for reach vehicle is recorded each minute, say. Such data are valuable to be used in real-time to estimate the current travel time on individual paths and links. However, the sparse structure of the data also introduces a measurement error, since it in general cannot be known which path the vehicle has taken. In this project we derive a consistent and computational feasible estimator to estimate a route choice model using sparse GPS data. Another feature of the proposed estimator is that it also addresses the problem of measurement errors related to the GPS position. In general, the GPS positions are recorded with stochastic errors, which also introduce a measurement err

Top page top