考虑前后多车的网联自动驾驶车辆混流跟驰模型

Owing to the proposal of new infrastructure construction strategies and the development of autonomous driving and connected network communication technologies, mixed-traffic flows, which consist of connected and automated vehicles(CAV), autonomous vehicles(AV), and regular vehicles(RV), are expected to exist for a long time. This paper proposes a mixed-flow following model for three types of vehicles-connected and automated vehicles, autonomous vehicles, and regular vehicles-considering the headway of multiple front and rear vehicles, the velocity and acceleration differences between multiple front vehicles and the host vehicle, and the relative distance from the host vehicle. In addition, numerical simulations involving typical scenarios were conducted. Brake and start the process of three types of mixed-flow numerical simulation results indicate that the model is feasible under several typical scenarios. The acceleration and velocity of the vehicle changed more gently. Results of the numerical simulations with different CAV ratios indicate that, the higher the CAV ratio of the fleet, the shorter is the time required for the overall fleet to recover to a stable state and the smaller is the fluctuation range. The numerical simulation results for the CAV homogeneous flow indicate that the unstable region of the MFRHVAD model is reduced by 33. 8% and the velocity fluctuation range of the fleet controlled by the MFRHVAD model is reduced by 14%, as compared with those when using the MHVAD model. Furthermore, the numerical simulation results for mixed CAV and AV flows indicate that the acceleration of the fleet controlled by the MFRHVAD model enters a relatively stable state 5. 5 s before that when using the PATH laboratory model. This model can be utilized for the queue control of homogeneous and mixed-traffic flows. In situations where it is difficult to perform actual vehicle experiments with mixed-traffic flows, this model can be applied for simulating the car-following behavior; hence, it can be helpful for road traffic management and developing infrastructure layouts for mixed-traffic flows.