Supplied by Wejo, this dataset contains GPS trajectories for around 50,000 vehicles during the month of July, representing over 1.8 million vehicle journeys and over 400 million individual records. An observation is available every 3 seconds on average during each journey.

The data contains a journey identifier, timestamp, longitude and latitude coordinates, as well as additional data fields for vehicle speed and bearing. The data is of high quality, with GPS records for every three seconds on average. This enables the successful implementation of map-matching algorithms as well as the identification of vehicle stops as well as periods of acceleration and deceleration.

This is one of the first times such a detailed and in-depth dataset detailing connected vehicle trajectories has been made available for academic research purposes.

Please see the data profile document for full details about the dataset.

If you are interested in applying to this dataset, please email CDRC Leeds Centre Manager Oli Mansell – o.j.mansell@leeds.ac.uk – in the first instance.