Evaluating the Safety and Mobility of the Cumulative-Anticipative Car-Following Model for Connected Autonomous Vehicles

Abstract

The advancements of vehicle automation are progressively improving resulting in safer driving environments in addition to more efficient mobility and fuel cost savings. However, autonomous and connected autonomous vehicles (AVs, CAVs) require decades to achieve complete market penetration. It is important to investigate the coexistence of conventional and autonomous cars during such a transition period. Traditionally, adaptive cruise control (ACC) and cooperative ACC (CACC) models were used for the AVs to guide their car-following. Recently, the cumulative-anticipative car-following (CACF) model was developed with consideration of the cumulative influences from surrounding vehicles through vehicle-to-everything (V2X) communication. This study further evaluates the safety and mobility performances of the CACF model for CAVs in mixed traffic through various sensitivity tests using the VISSIM simulation platform. The results demonstrate that the CACF model has promising improvements in roadway safety and network performances compared with the Wiedemann 99 and CACC models in mixed environments.