Dynamic Car-following Model With Jerk Suppression for Highway Autonomous Driving

Abstract

This study proposes a dynamic safe car-following strategy that is based on dynamic adjustment of headway time with jerk suppression. Reinforcement learning models trained with this strategy result in enhanced safety and driving comfort, validated using real driving data from the Next Generation Simulation (NGSIM) I-80 and HighD datasets. Simulation results demonstrate significant reduction in the risk of collisions. More importantly, low collision rates are maintained with driving speed profiles that are different from the training data, exhibiting cross-dataset generalizability. It also significantly improves driving comfort, with a 10% jerk reduction compared to existing models.