Skip to content
ITS Edulab > Projects > Project > In-vehicle sensor data for traffic management

In-vehicle sensor data for traffic management

Common current methods to measure traffic include induction loops, license plate recognition cameras, radar, laser and GSM. In the Netherlands, but also worldwide, induction loops are the most widely used traffic sensors. Induction loops however can be unreliable and not accurate enough for some applications. Traffic intensity and speed are only measured where loops are available. A typical distance between loop detectors on Dutch motorways is 500 meters. Video based monitoring can give more detailed information but also has its limitations and drawbacks, like costs and low detection rate in case of bad weather and at night.

The traffic data collection systems are used to monitor the traffic conditions on the road. This information can be used for traffic management applications like ramp metering, travel time prediction and dynamic speed limits. For some applications, like incident detection, the provided resolution of the induction loops is too low to adequately function. New traffic data collection systems which are able to provide data more frequently and with greater accuracy could enable possibilities for new traffic management applications and improve existing ones. In the Smart in-car test the CAN bus data of vehicles is gathered along with the GPS position of the vehicle and the acceleration of the vehicle in all three dimensions. The CAN (controller area network) bus is a specialized internal communication network that interconnects components inside a vehicle.

Combined with the GPS position, the CAN bus data can provide insight into many vehicle, driver, road and traffic characteristics. This research will mainly focus on the real time applicability of the CAN bus data for traffic management on motorways. Some examples of possible suitability of CAN bus data are: an abrupt steering manoeuvre can indicate an obstacle on the road, windscreen wiper usage can indicate adverse weather (e.g. rain) and activation of ABS might be an indication for a slippery road surface. This information can be used to warn drivers that are upstream of the incident location and to alarm the road authorities. Until now, there has not been a system that automatically and real time detects these traffic and road conditions. If the CAN bus data proves to be suitable for providing this sort of information, it can contribute to better traffic management and safer traffic. 



Start date: April 1, 2012
End date: March 11, 2013

Research topic:
Remaining topics

Research question:
Can data fusion improve traffic management?

Report:
Level of Service estimation with in-vehicle sensor floating car data