Evaluating Eco-driving Advice using GPS and CANBus data

Student thesis: Master thesis (including HD thesis)

  • Sabrine Conny Hjøllund Mouritsen
  • Karsten Jakobsen
4. term, Computer Science, Master (Master Programme)
Road transportation accounts for about 20% of the emission of CO2 in EU.
This can be reduced by driving more fuel efficient with the added benefit of lower expenditure on fuel.
Several advice on fuel efficient driving behaviour exists, but these are very abstract and lacks to be grounded in data.
A large set of real, high frequency GPS and CANBus data from four similar vehicles is used to evaluate multiple eco-driving advice.
CANBus data provides information about a vehicle such as the fuel consumption and the rounds per minute of the engine.
It is estimated that idling costs about 1 l/h in wasted fuel, and the fuel consumption per kilometer is 76 ml/km more when driving at 130 km/h compared to 80km/h.
Examples of stopping at traffic lights indicate an increase in fuel consumption of up to twice as much.
The four vehicles are compared and evaluated on their ability to follow the eco-driving advice.
There is a 22 % difference between the fuel efficiency of the drivers in this study.
Therefore there is potential for saving large quantities of fuel by informing and educating drivers.
No single eco-driving advice dominates, but multiple advice are found to have a positive effect on the fuel efficiency when followed.
Publication date7 Jun 2013
Number of pages20
ID: 77310273