In order to regulate traffic congestion under the condition when connected automated vehicles (CAVs) and human driving vehicles are mixed. Considering the difference of technical performance between CAVs and human driving vehicles, the macroscopic characteristics of mixed traffic flow and effects of CAV penetration level on it are discussed in this work. We proposed a concept of CAV platoon to explore both the internal and external car-following patterns. Based on classic improved car-following models, CAV nested car-following model is proposed and numerical simulation experiments are carried out. The results indicate that 1) external disturbance propagates backwards with time, and stop-time intervals and velocity oscillation increase with vehicle number; 2) the ability of mixed traffic flow in reducing disturbance and stabilizing the velocity oscillation improves either as CAV penetration level increases or as CAV platoons increase at the same CAV penetration level. Besides, results derived from this research study have positive practical significance for guiding the applications of CAVs and promoting the safety, stability and efficiency of road traffic system.

CICTP