VIA NOVA wins the "Best scientific paper" award

12 juni 2019

During the ITS European Congress that was held in the first week of June 2019 at Brainport Eindhoven, a research paper titled "Predicting Traffic Phases from Car Sensor Data" of the the project VIA NOVA won the "Best scientific paper" award. 

Cooperative mobility relies largely on (big) data already available in modern cars in terms of data from sensors, processors, navigation devices, cameras, etc. The problem is, however, using this data: the quality is not clear and even varies among car-brands and car-types. The project VIA NOVA investigates if modern vehicles could be used as sensors to monitor various conditions such as traffic flow, road conditions and weather conditions. The project will result in the development of a generic strategy and toolbox in which the specific quality of the data and needed quantity is related to the potential use within Cooperative Mobility. The outcome will enable users (talking traffic service providers, road administrators, app developers, traffic managers and even OEM’s) to judge whether data from cars can be useful in solving specific traffic-related use cases.

Here is the abstract of the paper published by the authors Emiliano Heyns, Shammy Unyal and Chris Huijboom.

"This research is an explorative study to look for the potential to predict traffic density from driver behavior using signals collected from the Controller Area Network(CAN) bus. The hypothesis is that driver behavior is influenced by traffic density in such a way that an approximation of the traffic density can be determined from changes in the driver behavior. Machine learning will be employed to correlate a selection of commonly available sensors on cars to the traffic density. Challenges in the processing of the data for this purpose will be outlined. This study is restricted to straight roads in order to isolate the steering behavior attributable to the traffic state influences rather than following the curve in the road. The results are encouraging that the correlation between driver behavior and traffic density can be established. An overall accuracy of over 95% is achieved with a precision of 92%. The recall rate however is low most likely caused by over-fitting due to the unbalanced data-set."

Bron: HAN Automotive Research