Largest multi-city traffic dataset publicly available.

Over the last four years, the Institute for Transportation Planning and Systems (IVT) at ETH Zurich and CITIES at NYU Abu Dhabi have gathered urban traffic data from over 40 cities worldwide, including London, Tokyo, and Zurich. In total, almost 5 billion vehicles covering a combined time span of 3.8 years were detected.

Now, it is our pleasure to announce the publication of our dataset for public research purposes.

Our “UTD19” dataset mainly consists of measurements from 22’820 loop detectors, making it the largest multi-city traffic dataset publicly available. Traffic is recorded using vehicle flow and occupancy (or speed) in relatively small aggregation intervals, typically 3-5min. In addition, we geo-coded the location of all detectors and the associated roads in WGS84 coordinates making map matching as easy as possible.