Collecting driving data to improve safety

Next year, 80 Volvo trucks in service at different fleets across Europe will be part of a large-scale research project aimed at making road transport safer and more efficient: the Large-Scale Field Operational Test on In-Vehicle Systems (euroFOT).

- This research project is fully in line with our ambition to build the safest trucks in the world, says Carl Johan Almquist, Safety Director at Volvo Trucks.

Comprehensive information
All trucks will be equipped with Adaptive Cruise Control including Forward Collision Warning, Lane Keeping Support and Fuel Efficiency Advisor systems. The test will help Volvo Group to evaluate these systems' impact on safety, mobility and environment. It will also show how drivers actually use the systems.

- Besides vehicle data gathered by sensors in and on the vehicle, we also collect visual information from four cameras. This gives us a good view of what happened in the cab and on the road at any given point in time, says Karsten Heinig, responsible for the Project at Volvo Technology.


Karsten Heinig

Unified way to collect data
Testing forms an integrated part of any technology development process. But a naturalistic driving study of this magnitude provides substantial advantages. Karsten mentions further development of the collision warning systems as one example:

-These systems rarely go off in real traffic. But with the massive amount of data collected in the field test, we should have the input required to optimize the algorithm on which the system is based to make it even more reliable.

Designing new hardware
However, to store and analyze the collected data imposes a challenge. 

- Trucks work hard in a variety of environments. The logger must endure extreme temperatures and be very resistant to shock and vibrations. We collaborate with Chalmers University of Technology and Volvo Cars to develop the hardware needed, says Karsten Heinig.

80 trucks collecting tons of data
Each truck will collect up to 1 gigabyte of data for every hour in service. And each vehicle is expected to be on duty for an average of 40 to 50 hours per week during the full year.

With 80 trucks monitored, more than 70 000 000 megabytes of data will be available when the field test is over.

Efficient analysis tools
Analysis of vehicle and sensor data, as well as some video analysis, will be automated. 

- We will use software to analyze the location of the driver's head and facing direction, as well as the pitch and roll of the driver's eyes, Karsten Heinig explains.

Video analysis is often crucial to identify potentially dangerous situations and maneuvers.

- Incidents and near crashes are very hard to detect using vehicle data only. So information regarding weather, traffic density and other factors revealed by the videos will be manually or semi-automatically coded, Karsten Heinig adds.

Cross-enterprise cooperation
A euroFOT network has been formed within the Volvo Group to make sure that the company gets as much knowledge as possible out of the field test. The set-up involves Volvo Technology, Volvo 3P, Volvo Trucks, Volvo Logistics, and Volvo Powertrain.

Also, the network relies on experiences from previous field operational tests carried out by Volvo Trucks North America and Mack Trucks.