Proposal of a Preventive Safety System Based on a Lane Change Probability Estimation Model for Motorcycle Riders

Abstract

In this study, we analyzed the effects of motorcycle riders' emotional states and human characteristics on their lane change behavior. As a result, we confirmed that motorcycle riders tend to change lanes when the valence level of Russell's emotion model is negative, even in risky situations with small time to collision (TTC) with the following vehicle in the adjacent lane. We also clarified indicators of human characteristics related to the lane change behavior of motorcycle riders. Finally, we constructed a model to estimate the lane change probability of a motorcycle rider using their emotional states, human characteristics, and surrounding vehicle information based on logistic regression analysis in order to propose a cooperative Human-Machine Interface (HMI) for two-wheeled and four-wheeled vehicles.