Towards a Review on Simulated ADAS/AD Testing

Abstract

Vehicle and traffic simulation is a common practice for testing and evaluating advanced driver-assistance systems (ADAS) and autonomous driving (AD). As a result, the literature mentions numerous simulators capable of simulating ADAS/AD implementations. In this study, we investigate previous surveys that cover multiple scenarios and initiate a systematic review targeting simulators for testing ADAS/AD. Our results show that the literature mentions, in total, 181 simulators capable of evaluating one or more ADAS/AD implementations. Furthermore, according to previous surveys and reviews, the most popular simulators are CARLA, Airsim, and SUMO. Finally, our results uncover that every five years, the number of novel simulators added to the literature grows at least quadratically, showing that further review is necessary to address the differences between these simulators and understand the simulator landscape from an ADAS/AD testing perspective.