Evaluation of Safety Effects of Driver Assistance Systems Through Traffic Simulation

Road safety is a major concern in all countries, and large effortsare constantly dedicated to create safer traffic environments. Todayincreasing attention is turned toward active safety improvingcountermeasures that improve road safety by reducing accidentrisks. Such active countermeasures include advanced driver assistancesystems (ADAS). To ensure that these new applicationsresult in real safety improvements, a priori estimations of safetyeffects are needed. This paper considers estimation of the safetyeffects of ADAS through traffic simulation. Requirements imposedon a traffic simulation model to be used for ADAS evaluation arepresented, and a car-following model to be used in simulations thatinclude ADAS-equipped vehicles is proposed. ADAS have an impacton traffic through the functionalities of ADAS and throughchanges in driver behavior for ADAS-equipped vehicles. Driverbehavior for ADAS-equipped vehicles has usually not been consideredin previous simulation studies, including those for ADASequippedvehicles. Simulation runs of rural road traffic that usedthe proposed car-following model did, however, indicate that behavioralchanges caused by the ADAS were important factors forthe safety impact. Modeling of the behavior of drivers in ADASequippedvehicles is therefore essential for reliable conclusions onthe road safety effects of ADAS.