基于驾驶行为生成机制的智能汽车类人行为决策  

Human-Like Behavior Decision-Making of Intelligent Vehicles Based on Driving Behavior Generation Mechanism

作　　者：宋东鉴 朱冰[1] 赵健[1] 韩嘉懿[1] 刘彦辰[1] Song Dongjian;Zhu Bing;Zhao Jian;Han Jiayi;Liu Yanchen(Jilin University,State Key Laboratory of Automotive Simulation and Control,Changchun 130022)

机构地区：[1]吉林大学,汽车仿真与控制国家重点实验室,长春130022

出　　处：《汽车工程》2022年第12期1797-1808,共12页Automotive Engineering

基　　金：国家自然科学基金(52172386)资助。

摘　　要：本文通过分析驾驶人驾驶行为生成机制,构建了类人行为决策策略(HBDS)。它具有匹配驾驶行为生成机制的策略框架,通过最大熵逆强化学习得到类人奖励函数,并采用玻尔兹曼理性噪声模型建立行为概率与累积奖励的映射关系。通过预期轨迹空间的离散化处理,避免了连续高维空间积分中的维数灾难,并基于统计学规律和安全约束对预期轨迹空间进行压缩和修剪,提升了HBDS采样效率。HBDS在NGSIM数据集上进行训练和测试的结果表明,HBDS能做出符合驾驶人个性化认知特性和行为特征的行为决策。In this paper,a human-like behavior decision-making strategy(HBDS)is established by analyzing drivers’ driving behavior generation mechanism. HBDS has a framework that matches the driving behavior generation mechanism,obtains the human-like reward function through maximum entropy inverse reinforcement learning,and adopts the Boltzman noisily-rational model to build the mapping relationship between behavior probability and its cumulative reward. By discretizing the expected trajectory space,the curse of dimensionality in the integration of continuous high-dimensional space is avoided,and based on statistical law and safety constraint,the expected trajectory space is compressed and pruned,enhancing the sampling efficiency of HBDS. The strategy is trained and tested on NGSIM dataset,and the results show that HBDS can make behavior decisions that conform to the driver’s personalized cognitive and behavioral characteristics.

关 键 词：智能汽车 类人驾驶 行为决策 逆强化学习 

分 类 号：U463.6[机械工程—车辆工程]