Modeling lateral interferences between motor vehicles and non-motor vehicles: A survival analysis based approach

Mixed traffic consisting of motor vehicles and non-motor vehicles has a significant impact on traffic efficiency and safety. Aiming at analyzing the interaction between motor vehicles and non-motor vehicles, a lateral interference model based on survival analysis approach is proposed. The trajectory data for motor vehicles and non-motor vehicles are acquired from video capture and processed with coordinate conversion. The lateral interference between the motor vehicle and the non-motor vehicle is analyzed by the relation between reference probability and lateral distance. The shortest distance which is able to cause lateral interference is defined as critical distance and will be discussed quantitatively. The results show motor vehicle drivers and non-motor vehicle drivers endow different sensitivities to the lateral interference. The motor vehicle drivers are inclined to keep longer lateral distance to non-motor vehicles, while the non-motor vehicle drivers accept shorter lateral distance. Also the sensitivity of bicycle riders in avoiding lateral interference is greater than that of the electric bicycle riders. This study attempts to provide a more comprehensive and profound understanding of the safety behavior in mixed traffic circumstances.