Markov probabilistic decision making of self-driving cars in highway with random traffic flow: a simulation study

作     者：Yang Guan Shengbo Eben Li Jingliang Duan Wenjun Wang Bo Cheng 

作者机构：Tsinghua UniversityBeijingChina Department of Automotive EngineeringTsinghua UniversityBeijingChina 

出 版 物：《Journal of Intelligent and Connected Vehicles》 (智能网联汽车（英文）)

年 卷 期：2018年第1卷第2期

页      面：77-84页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：Markov decision process Decision-making Dynamic programming Self-driving cars 

摘      要：Purpose–Decision-making is one of the key technologies for self-driving *** high dependency of previously existing methods on human driving data or rules makes it difficult to model policies for different driving ***/methodology/approach–In this research,a probabilistic decision-making method based on the Markov decision process(MDP)is proposed to deduce the optimal maneuver automatically in a two-lane highway scenario without using any human *** decision-making issues in a traffic environment are formulated as the MDP by defining basic elements including states,actions and basic *** and reward models are defined by using a complete prediction model of the surrounding *** optimal policy was deduced using a dynamic programing method and evaluated under a two-dimensional simulation ***–Results show that,at the given scenario,the self-driving car maintained safety and efficiency with the proposed ***/value–This paper presents a framework used to derive a driving policy for self-driving cars without relying on any human driving data or rules modeled by hand.