摘要
提出了考虑驾驶人驾驶习性的自适应车道偏离预警策略.通过实车驾驶数据采集平台采集驾驶人的驾驶行为数据,并基于模糊聚类对驾驶数据进行聚类处理,进而利用广义回归神经网络(GRNN)模型实现了驾驶人驾驶习性辨识策略;建立车道偏离时间估算模型,设计个性化的车道偏离预警系统;最后,通过驾驶模拟器进行测试验证.结果表明,所提出的考虑驾驶人驾驶习性的自适应车道偏离预警策略能够在有效辨识驾驶人驾驶习性的基础上,提高车道偏离预警的适用性.
An adaptive lane departure warning strategy considering driver’s driving style is proposed.Firstly,the driver’s driving style data are collected from the real vehicle driving data acquisition platform,and the driving data are clustered based on the fuzzy clustering.Then,the driver’s driving style identification strategy is established by using the generalized regression neural network model.Secondly,the estimation model of lane departure time is established,and the individualized lane departure is designed.Finally,the departure warning system is validated by a driving simulator.The results show that the proposed strategy could improve the applicability of lane departure warning based on the effective identification of driver’s driving style.
作者
朱冰
李伟男
赵健
韩嘉懿
ZHU Bing;LI Weinan;ZHAO Jian;HAN Jiayi(State Key Laboratory of Automotive Simulation and Control,Jilin University,Changchun 130022,China;Key Laboratory of Bionic Engineering of the Ministry of Education,Jilin University,Changchun 130022,China)
出处
《同济大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2019年第S01期171-177,共7页
Journal of Tongji University:Natural Science
基金
国家重点研发计划(2016YFB0100904)
国家自然科学基金(51775235,U1564211)
吉林省自然科学基金(20170101138JC).
