A Personalized Human-Like Lane-Changing Trajectory Planning Method for Automated Driving System

Abstract

This article describes a human-like automated lane-changing system to mimic the lane-changing maneuver of human drivers in order to make automated vehicles more realistic. In this system, both the personalized factors of human drivers and the traffic environmental factors are considered to derive a personalized human-like lane-changing trajectory planning model, which includes a longitudinal driving behavior model and a lateral lane-changing trajectory planning model. The longitudinal model is derived based on the traffic environment information to predict the longitudinal speed of human-driven vehicles in the lane-changing processes, and also to analyze the longitudinal driving habits of drivers in different driving scenarios. The lateral model plans a suitable lateral lane-changing trajectory preferred by the driver. The personalized parameters of the longitudinal and lateral models are calibrated by the driving data with multi-dimensional time series regression method. The simulation results show that the proposed method can well realize the human-like and personalized lane-changing trajectory planning.