A Car-following Model Considering Acceleration of Multiple Vehicles for Mixed Traffic Flow

<div class="section abstract"><div class="htmlview paragraph">The practice of vehicle platooning for managing mixed traffic can greatly enhance safety on the roads, augment overall traffic flow, and boost fuel efficiency, garnering considerable focus in transportation. Existing research on vehicle platoon control of mixed traffic has primarily focused on using the state information of the leading or head vehicle as control input for following vehicles without accounting for the driving variability of Human-driven Vehicles (HDVs), which does not conform to the driving conditions of vehicles in reality. Inspired by this, this paper presents a car-following model for Connected and Automated Vehicles (CAVs) that utilizes communication with multiple preceding vehicles in mixed traffic. The study further investigates the impact of parameters such as the speed and acceleration of preceding vehicles on the car-following behavior of CAVs, as well as the overall effect of different CAV penetration rates on mixed traffic flow. Firstly, a mixed-vehicle platoon model is constructed, and an improved multi-vehicle-following topology controller is proposed. Secondly, based on the multi-leading-vehicle communication topology, the head-to-tail transfer function of the vehicle platoon is derived, and the impact of the communication topology on platoon stability is analyzed under different CAV penetration rates. Finally, the proposed vehicle-following model is verified on the SUMO simulation platform. The experimental results demonstrate that, compared to the Intelligent Driver Model(IDM), the proposed model exhibits more minor speed fluctuations and superior following efficiency. Additionally, under the proposed car-following model, the stability of mixed traffic flows can be enhanced as the penetration rate of CAVs increases. This research provides theoretical and technical support for vehicle platoon control issues in mixed-traffic environments.</div></div>