Floating Car Data


Wikipedia describes Floating car data (FCD) al follows:

Floating car data (FCD), also known as floating cellular data, is a method to determine the traffic speed on the road network. It is based on the collection of localization data, speed, the direction of travel and time information from mobile phones in vehicles that are being driven. These data are the essential source for traffic information and for most intelligent transportation systems (ITS). This means that every vehicle with an active mobile phone acts as a sensor for the road network. Based on these data, traffic congestion can be identified, travel times can be calculated, and traffic reports can be rapidly generated. In contrast to traffic camerasnumber plate recognition systems, and induction loops embedded in the roadway, no additional hardware on the road network is necessary.”

Moreover, floating car data can also be captured by navigation systems such as PNDs (Personal Navigation Devices) and Telematics systems. Floating car data is used by applications like google maps and Waze to predict travel times. Floating car data typically has a location and a timestamp.

Sometimes an identifier of the probe (mobile phone, car navigation system) capturing the data is stored as well. In that case, it is necessary to take the steps as defined by the new European regulation GDPR, to ensure privacy protection. More precisely, it’s essential to make sure the information cannot be linked back to the individual car or person generating the data.

PoliVisu works with this type of data in the framework of Issy-les-Moulineaux pilots, more particularly with data of the Paris Region. Floating car data is typically data collected by private market players. The data used in PoliVisu was obtained from Mediamobile.

When best to use
Floating car data enables traffic information:
  • Traffic jams by determining the speed driven versus the freeflow speed (speed not slowed down because of traffic, typically measured at night);
  • Relative nr of probes on the street: Since no floating car dataset has information about all traffic, the absolute number of probes cannot be derived.  When the sample is relevant, the relative number of probes per travelled road segment can be derived;
  • The direction of the traffic flow: In what direction are the probes (measured vehicles) driving on the network. This enables checking whether the direction of traffic flow in a reference dataset is correct;
  • Measuring traffic flows on the network: Enable to get a view on the mobility patterns in the city;
  • Turn restrictions: Verify turn restrictions in reference datasets versus turns that were made by the probes;
  • Speed limits: Verify speed restrictions in reference datasets versus speeds driven by probes.
The added value and strength of floating car data lies in the granularity and not in the completeness of the data. It is difficult to get a clear view of the volume of FCD that is necessary to have accurate measurements. But it looks that about 10-15% coverage leads to acceptable and detailed traffic information.
Main data elements:

The data used in the Issy contains these “typical” data elements:

  • Road Network data: Geometry, road classifications, length, free-flow speed;
  • Vehicle Position data: Vehicle ID, location, timestamp;
  • Vehicle Traces: Positions projected onto the links;
  • Observations: Vehicle id, link id, coverage (completely travelled or not) timestamp, speed;
  • Speed: Link id, average speed, timestamp

Leave a Review.

You must be logged in to post a comment.