Paper Title
Lane-Changing Trajectory Prediction Based on Adaptive Fuzzy Reinforcement Learning

Abstract
Lane change is not only one of the significant impact on traffic safety and efficiency, but also a challenge with autonomous vehicle (AV) technology. Through evaluating and considering the surrounding environment such as surrounding vehicles, pedestrians, the process of changing lanes is complicated and requires high precision. In this paper, a novel of lane-changing trajectory prediction strategy is designed to support AV in the lane change operation. Adaptive fuzzy reinforcement learning (AFRL) is constructed to determine the lane change trajectory respect to the equidistant sampling of lane change time. Therefore, the safety trajectories are calculated so that the host vehicle can easily choose. Several simulation results demonstrate that the proposed approach has outstanding performance in predicting lane change trajectories. Keywords - Adaptive Fuzzy Neuro Network, Lane-changing Trajectory Prediction, Reinforcement Learning