Laser-and-vision based probe car system toward realtime lane-based traffic data collection

A more functional and intelligent traffic control system needs not only static information about road, but also real-time dynamic information such as volume, speeds, queue lengths, and incident occurrences. Due to this reason, collecting traffic data methods have been evolving considerably and the access to real-time traffic data is becoming routine worldwide. The traditional collection methods using roadside sensors are necessary but not sufficient because of their limited coverage and expensive costs of implementation and maintenance. To solve the limit of roadside sensors, the popular probe car system can collect real-time traffic data from in-vehicle devices through mobile phones or GPS, BUT only road-link based information can be collected due to poor GPS signal in urban areas. In addition, traffic data (e.g. speed, density of vehicle) surround vehicle are not captured at all, because only the location of probe car itself is observed. In this paper, we consider localization and surround traffic data collection together to develop an advanced probe car system with various vehicle sensors, which can get not only more precision localization with multi-sensor-based positioning system, but also traffic information surround vehicle coming from and mining from multiple sensors.