Abstract:
Data generated from telemetry devices, such as acceleration and speed, are used in a variety of industries to determine the risk profile of a driver. This paper considers the addition of road gradient as a contextual variable to a driving behavior model to determine specifically if the risk of a driver is influenced by different road grades. This behavioral risk is demonstrated by comparing the performance of heavy vehicle drivers with and without the addition of road grade as a variable. This is done by using accelerometer data of 48 heavy goods vehicles over an entire month, and appending elevation data from digital elevation model (DEM) data, specifically that of the Shuttle Radar Topography Mission (SRTM), to the vehicles’ GPS traces. The elevation values are used to calculate road grade values which are then categorized in five grade levels. The results show that there is a clear influence of gradient on the behavior of the drivers studied. There is also evidence that steeper sections on a road can cause a change in an individual’s performance when compared with the group.
