This paper presents a method for evaluating the effectiveness of a forward-looking bicyclist safety system for passenger cars. The method starts with a detailed analysis of 4286 bicycle-to-car accidents recorded between 2000 and 2013 in the German in-depth accident database GIDAS. The results of the analysis identify relevant parameters influencing bicycle-to-car-accidents which include surrounding conditions (time and location of the accident, obstruction), vehicle parameters (collision speed) and cyclist parameters (helmet use, cyclist age). An aggregation of the accidents to reference scenarios is subsequently presented. With more than 60 % of all bicycle-to-car accidents a crossing cyclist is the major scenario followed by accidents between cyclists and turning vehicles and accidents in longitudinal traffic. A further investigation of these scenarios is conducted by generating plots for the trajectories of the cyclists relative to the vehicle. Results so far for the overall method are presented. Further steps to be taken are described. This includes deriving test scenarios of the reference scenarios, set up a risk-based assessment of the safety system, carry out a benefit assessment with simulations and the evaluation of the risk-based assessment with the simulative benefit assessment. The complete study contributes to a deeper understanding of cyclist accidents and it delivers a method for the assessment of active safety systems preventing bicycle-to-car accidents. (Author/publisher)
Abstract