Vehicle-Following Model Using Virtual Piecewise Spline Tow Bar

Vehicle following has always been a hotspot of research in the transportation system. It can be used in many applications, such as unmanned ground vehicle (UGV) queue marching, traffic congestion relief, automatic driving systems, etc. Trajectory tracking has been the main consideration in previous studies, but the research about following distance is rare. In this paper, a novel vehicle-following model is proposed for independent vehicle following. In the proposed model, multisensor information is fused to promote the detection accuracy. Moreover, a virtual piecewise spline tow-bar is built between the leader vehicle and the follower vehicle. According to the state of virtual tow-bar, the following speed and heading can be well controlled, to keep a reasonable gap between two vehicles. The proposed vehicle-following model is designed for low speed and minor vehicle spacing case. By using the proposed vehicle-following model, the performance of multple-vehicle automatic cruise can be improved, resulting in a steady traffic flow on an express way. Then the stop-and-go waves of congestion can probably be relieved. Simulation and experiment are conducted to validate the efficiency and robustness of the proposed vehicle-following model.