Connected Car 2017-08-11T09:59:27+00:00

Connected Car

We see and treat the car as a sensor platform that, connected with other cars, bares huge potential to increase driver safety and sustainability!

Therefore, we developed a plug-and-play system for the car that is able to read and process detailed data from the car’s Controller Area Network (CAN) Bus.

We use this data to investigate four main areas of connected mobility:

  • Eco-Driving – the impact of realtime and contextual information on driver behaviour
  • Road Safety – the ability to detect dangerous sections of road from driving data, and warning drivers as they approach these
  • Driver Identification – can we identify individuals based on their driving behaviour, i.e. how hard they brake
  • Driver Health – are we able to predict the stress levels of the driver based on how they are driving, such as a higher frequency of aggressive manoeuvres

In order to tackle these challenging topics we have several field tests in cooperation with Touring Club Swiss (TCS), where professional drivers test our system and interventions. Watching the TV show feature (in German) below will help give some more insights on the aspects of our projects (beginning at 1:25):


Current Projects


Almost 17% of the worldwide CO2-emissions can be ascribed to road transportation. Using information systems (IS)-enabled feedback has shown to be very efficient in promoting a less fuel-consuming driving style. Today, in-car IS that provide feedback on driving behaviour are in the midst of a fundamental change. Increasing digitalization of in-car IS enable virtually any kind of feedback. Still, not much is known about the impact such systems can have in real life and how to leverage this potential. As such, we aim to address the need for rigorous research on the impact of eco-driving feedback in our unique field test setting. 


Road Safety

Despite continuous investment in road and vehicle safety, as well as improvements in technology standards, the total amount of road traffic accidents has been increasing over the last decades. Consequently, identifying ways of effectively reducing the frequency and severity of traffic accidents is of utmost importance. Going beyond existing research, we developed a comprehensive in-vehicle decision support systems, which provides accident hotspot warnings to drivers based on location analytics applied to a Swiss historical accident dataset, composed of over 266,000 accidents. The system was tested in a country-wide field test of 57 professional TCS drivers, with over 170,000km driven during a four-week period. Ultimately, we demonstrate that in-vehicle warnings of accident hotspots have a significant improvement on driver behaviour over time.

Driver Identification

In an ever-more digitalised world, the effortless and secure identification of a person has become a key enabler of many new products, services and business models. However, person identification today is often cumbersome and insecure. Consequently, there is a new generation of frictionless and reliable identification processes in development that apply machine learning techniques such as face or voice recognition to authenticate users. In light of these recent developments, we explore whether the approach of machine-learning-based person identification can be applied to the car.

This graph shows the fraction of correct identification of 15 drivers in dependence of the number of feature events using a brake-event-detection approach

Driver Health – Under Construction!

This is our newest topic and currently being investigated, stay tuned for more details!


Get in touch!

André Dahlinger
André DahlingerPh.D. candidate and doctoral researcher

Benjamin Ryder
Benjamin RyderPh.D. candidate and doctoral researcher

Bernhard Gahr
Bernhard GahrPh.D. candidate and doctoral researcher