Car-Following Model of ULVs using a Deep Learning Model

In recent years, various types of vehicles have been developed and made available for the road. This study clarifies the driving characteristics of Ultra Lightweight Vehicles (ULVs) that consume less energy. First, coordinate data in units of 0.1 seconds are created by observing traffic flows including ULVs using unmanned aerial vehicles (UAVs). In addition, a car-following model is developed using the long short-term memory (LSTM) model, which has attracted attention as a deep learning model. Consequently, a model with a higher degree of fitness than a non-linear model is developed.