Human-Like Decision Making for Automatic Lane Changes at Congested Highway On-Ramp

This study considers a merging scenario on a congested highway on-ramp and develops a system that autonomously performs lane changes from an on-ramp to the main lane. In particular, we focused on constructing a method for selecting a target space in a congested traffic flow into which the host vehicle enters itself. Driving-behavior data of human drivers were collected using a driving simulator, and the proposed method was constructed by mimicking human decision-making mechanisms through machine learning. This enables autonomous vehicles to make natural decisions that are similar to those of human drivers, thereby enhancing the social acceptance of autonomous driving technologies. Simulation experiments confirmed that the proposed method ensures sufficient safety while selecting a target space comparable to that chosen by human drivers.