Research on vehicle driving characteristics in narrow lanes based on different vehicle-following states

Drivers could be more careful when vehicles move in the non-standard narrow lane. This might cause the drivers to behave differently from other normal-size lanes. To investigate the characteristics of the traffic flow in narrow lanes, this paper collected the vehicle trajectory data on four-lane road segments with 2.8m width and 3.25m width respectively. Firstly, the spatial distribution of wheel tracks in the lane under low traffic volume is studied, and the results show that the wheel trails’ distribution is obviously affected by three factors: speed, lane width, and lane in cross-section position, which are significantly different from normal-size lanes. Secondly, statistical analysis shows that vehicle-following status in narrow lane environment can be divided into three types: staggered following, parallel following, and transitional following. Finally, a multivariate logistic regression model is constructed to predict the driver’s following strategy under different driving micro-environment parameters. The research results can support the fine management of lanes and help to improve the driving strategy of autonomous vehicles.