Using vehicle fleet data to quantify the risk of critical driver surprise in traffic

Important results from the project

Given the overall results, we believe that this pilot project has met its objectives. We have successfully conducted several analyses of driver behavior-related parameters in large-scale vehicle data. It also proved possible to create a meaningful design for crowd-sourcing studies of comfort zone boundaries in a way that allows for the collection of relatively large data sets at a faster pace than can be achieved with traditional local user studies.

Expected long term effects

For the project partners, this is in many ways the first step into a new kind of research environment where engineering judgement can be substantially enhanced with large sets of relevant empirical data from either the customer fleet or from global large scale data collection via crowd-sourcing. For safety systems, this means a completely new approach and way of working for functional development, the importance of which is hard to overestimate.

Approach and implementation

The project started out as planned but then suffered an early setback when key personnel left Volvo Cars for other missions. Finding a good replacement took some time and led to us having to extend the time frame of the project, but once the new personnel was in place, the project has proceeded according to plan, and the intended goals have been obtained. The collaboration between the project partners has worked very well.