Autonomous vehicles can not only ease traffic congestion, improve road safety, but also better adapt to the crowd. In comparison, today's traffic congestion, frequent traffic accidents and other issues have become increasingly prominent, hence the impact of autonomous vehicles on traffic flow has become a research hotspot. Generally, Autonomous vehicles join the traffic flow in the form of vehicle formations, and even in the future, dedicated lanes will be set up for self-driving cars. However, after the self-driving car enters the traffic flow, its influence on the traffic flow is unknown. Therefore, this paper studies the effects of PLOEG and CACC car-following models and different autonomous vehicle platoon on traffic flow. Through the two-way coupling simulation experiment, the auto-driving car uses different car-following strategies to test traffic capacity. The experimental results show that the PLOEG strategy is slightly better than the CACC strategy for both speed modes in a specific scenario; The traffic flow of the two strategies has nothing to do with the size of the platoon, it has a relationship with the headway time. The smaller the headway time, the larger the traffic capacity.

CICTP