Publication

The majority of OEMs and system suppliers are developing detection sensors for crash avoidance and crash mitigation systems in the pre-crash phase. Methods are however lacking for testing the performance of the detection sensors. In scenariobased testing, all driver-vehicle-environment parameters are defined for a number of normative accident scenarios, which should be addressed, regardless of system technology. In the US, test scenarios have been defined for integrated vehicle-based safety systems (IVBSS) that alert the driver of a light vehicle or a heavy truck to an impending crash. The test scenarios have been suggested for rear-end, lane change, run-off-road crashes, which have been ranked highest based on frequency. Close behind in rank are intersection related crashes but for which test scenarios are still to be defined. The present study compiles and analyses statistics regarding intersection crashes from previous studies and three in-depth accident databases. The objective is to rank intersection crash scenarios on the basis of accident frequency and injury severity as well as to find crash details to support the specification of test scenarios. The focus of the present study is on urban crossing path crashes involving light vehicles, i.e. vehicles ranging from passenger cars to light trucks. The characteristics of the identified vehicle paths are compared with paths and velocity profiles extracted from a study of two monitored intersections. The results showed that three crash scenarios were most frequent and ranked in the following order; Straight Crossing Paths (SCP), Left Turn Across Path - Opposite Direction (LTAP/OD) and Left Turn Across Path - Lateral Direction Conflict (LTAP/OD) (Table i). The results indicated that the majority of crashes had occurred in unsignalised four-way intersections with a posted speed limit of 35 mph or 50 kph. Based on these findings and additional details regarding approach speeds and manner of collision, three test scenarios are suggested for detection sensors in urban intersections.