Personalized Autonomous Driving Using Model Predictive Control

-Parameter Estimation of Personalized Cost Function-

Abstract

This paper presents an autonomous vehicle control that reflects individual driving characteristics based on the model predictive control with personalized cost function. To realize this autonomous driving, driving data of various drivers are collected through experiments on a driving simulator (DS) at first. Secondly, parameters of cost function of model predictive control are estimated so as to reproduce the characteristic of individual driver. Finally, the proposed method is implemented on DS and the validity of the proposed method is verified through experiments.