Comparative analysis on information fusion conflict redistribution and driving risk classification

Abstract

In this paper, risk level correlation and classification based on the recorded driving activities' data about sharp maneuvering tasks ensuing with the human being who is controlling the technical system, i.e., the car, in traffic is presented. The dataset is constituted by time stamped and geographically referenced driving maneuver information, which is exceptionally occurring when exceeding a given threshold acceleration in both longitudinal and lateral direction and a speed limit given as the static attribute of the road map data. Ground truth risk level is identified in terms of the change in vehicle collision property damage cost for the analyzed time period. Distribution of conflict masses generated by harsh driving activities and overspeeding is studied in order to predict the risk level about making an accident. High risk group is identified by PCR6 method with a level of accuracy of 80%.