Performance of Autonomous Vehicles in Mixed Traffic under different Demand Conditions

Abstract

Autonomous Vehicles (AVs) are considered one of the potential solutions to future urban mobility with several promised benefits regarding safety and traffic operation. Despite of expected benefits, these vehicles will take decades to have full market penetration and before that, AVs will co-exist with Conventional Vehicles (CVs), which may affect the performance of AVs owing to different driving logic than CVs. The aim of this study is to quantify the impacts of varying penetrations of AVs when introduced in mixed traffic conditions. The study employed simulation environment VISSIM to study the different scenarios based on the percentage of AVs in mixed traffic, category of AVs and varying demand levels. The findings show that at lower demand levels (1000 veh/hr and 2000 veh/hr), CVs and three categories of AVs produced similar results. However, cautious and normal AVs negatively affect traffic operations when the demand level is increased. At demand-3 (3000 veh/hr), the penetration rates of cautious AVs greater than 50% shows negative impact on performance. At demand-4 (4000 veh/hr), even a small proportion (25%) of cautious AVs can negatively affect performance, and a similar effect is observed for normal AVs with a penetration rate greater than 75%. For speed, the minimum reduction with the increase in demand is observed for aggressive AVs, followed by conventional vehicles, normal AVs and cautious AVs. It can be concluded that the aggressive AVs produced better delays, queue length, speed and conflicts than CVs, cautious AVs and normal AVs at the highest demand levels.

Investigation of Object Detection and Identification at Different Lighting Conditions for Autonomous Vehicle Application

“…However, improvement is needed so that the method can perform well for high distance assessment and in conditions when it is night and rain. Studies conducted by [23][24][25] have also emphasized the significance of considering the visibility of object detection in relation to weather conditions and the environment.…”