Estimating traffic conflict severity for Connected and Automated vehicles using simulation-based surrogate safety indicators

Abstract

Connected and Autonomous Vehicles (CAV) is the developing summit of the integration between artificial intelligence (AI), robotics, automotive design and information technologies. Many researchers are investigating their effects on traffic safety. This study tries to quantify the volume of incidents when sharing the road human-driven vehicles and fully CAV. After modeling the geometry of 4.5 km of motorway and the parameters of connectivity and automation using Aimsun Next platform, several scenarios of the percentages of CAV (0%, 25%, 50%, 75%, and 100%) were driven in microsimulation runs. Then the microsimulation generated vehicles trajectories that are used to identify conflicts using the Surrogate Safety Assessment Model (SSAM). The results of this analysis confirm previous research in that the reduction of number of conflicts will be up to 35% with low and moderate penetration rates of CAV and more than 80% if the road is operated only with CAV.