驾驶员转向特性分类与在线辨识研究
|
摘要
针对汽车驾驶员转向特性分类与在线辨识问题,基于驾驶模拟器实验台,设计典型实验工况,选取具有一定驾驶经验的驾驶员进行实验研究。采集驾驶员转向过程实验数据并提取特征值,应用模糊C均值聚类算法将驾驶员转向特性分为谨慎型、一般型和激进型三类;根据分类实验数据应用BP神经网络算法建立驾驶员转向特性辨识模型,对模型辨识准确度进行离线验证,并基于辨识模型设计在线辨识方法。实验结果表明:研究方法能够实现对驾驶员转向特性的合理分类和在线准确辨识。
For the problem of classification and identification of steering characteristics on-line of vehicle drivers,the experiment study is carried out based on the driving simulator experiment platform,designing typical experimental conditions and selecting experienced drivers in this paper. The experimental data of the driver's steering process is collected and the characteristic values are extracted. The fuzzy C means clustering algorithm is used to divide the driver's steering characteristics into three types including cautious type,general type and radical type. According to the classification experimental data,the model of driver's steering characteristic identification is established by application of BP neural network algorithm. The accuracy of model identification is verified off-line. On-line identification method is designed based on the identification model. The experimental results show that the study method can realize the classification reasonably and online identification accurately for the steering characteristic of the drivers.
For the problem of classification and identification of steering characteristics on-line of vehicle drivers,the experiment study is carried out based on the driving simulator experiment platform,designing typical experimental conditions and selecting experienced drivers in this paper. The experimental data of the driver's steering process is collected and the characteristic values are extracted. The fuzzy C means clustering algorithm is used to divide the driver's steering characteristics into three types including cautious type,general type and radical type. According to the classification experimental data,the model of driver's steering characteristic identification is established by application of BP neural network algorithm. The accuracy of model identification is verified off-line. On-line identification method is designed based on the identification model. The experimental results show that the study method can realize the classification reasonably and online identification accurately for the steering characteristic of the drivers.
引文
[1]Farid M N,Kopf M,Bubb H.Methods to develop a driver observation system used in an active safety system[J].VDI Berichte,2006,1960:639-650.
[2]Deng Wei-wen.Electrification and intelligence-technologies that drive future vehicles[J].Automotive Safety and Energy,2010,1(3):179-189.
[3]D Filev,J Lu,K Prakah-Asante.Real-time driving behavior identification based on driver-in-the-loop vehicle dynamics and control[C].Proceedings of the 2009 IEEE International Conference on Systems,Man,and Cybernetics,2009.
[4]J Lu,D Filev,K Prakah-Asante.From vehicle stability control to intelligent personal minder:real-time vehicle handling limit warning and driver style characterization[C].Computational Intelligence in Vehicles and Vehicular Systems,2009.
[5]MCG Quintero,JO Lopez,ACC Pinilla.Driver behavior classification model based on an intelligent driving diagnosis system[C].2012 15th International IEEE Conference on Intelligent Transportation Systems,2012.
[6]Pentland A,Liu A.Modeling and prediction of human behavior[J].Neural Computation,2009(11):229-242.
[7]A Wahab,TG Wen,N Kamaruddin.Understanding driver behavior using multi-dimensional CMAC[C].6th International Conference on Information,Communications&Signal Processing,2007.
[8]王建强,迟瑞娟,张磊.适应驾驶员特性的汽车追尾报警-避撞算法研究[J].公路交通科技,2009,1(10):22-28.(Wang Jian-qiang,Chi Rui-juan,Zhang Lei.Study on forward collision warning avoidance algorithm based on driver characteristics adaptation[J].Journal of Highway and Transportation Research and Development,2009,1(10):22-28.)
[9]Wang J,Lu M,Li K.Characterization of longitudinal driving behavior by measurable parameters[J].Transportation Research Record:Journal of the Transportation Research Board,2010,21(1):15-23
[10]吕岸,胡振程,陈慧.基于高斯混合隐马尔可夫模型的高速公路超车行为辨识与分析[J].汽车工程,2010,32(7):630-634.(Lv An,Hu Zhen-cheng,Chen Hui.Recognition on analysis on highway overtaking behavior based on gaussian mixture-hidden markov model[J].Automotive Engineering,2010,32(7):630-634)
[11]吴坚,姚琳琳,朱冰.基于高斯混合模型的驾驶员个人特质辨识[J].吉林大学学报:工学版,2015,45(1):38-43.(Wu Jian,Yao Lin-lin,Zhu Bing.Identification of driver individualities using gaussian mixture model[J].Journal of Jilin University:Engineering and Technology Edition,2015,45(1):38-43.)
[12]张磊,王建强,杨馥瑞.驾驶员行为模式的因子分析和模糊聚类[J].交通运输工程学报,2009,9(5):121-126.(Zhang Lei,Wang Jian-qiang,Yang Fu-rui.Factor analysis and fuzzy clustering of driver behavior pattern[J].Journal of Traffic and Transportation Engineering,2009,9(5):121-126.)
[2]Deng Wei-wen.Electrification and intelligence-technologies that drive future vehicles[J].Automotive Safety and Energy,2010,1(3):179-189.
[3]D Filev,J Lu,K Prakah-Asante.Real-time driving behavior identification based on driver-in-the-loop vehicle dynamics and control[C].Proceedings of the 2009 IEEE International Conference on Systems,Man,and Cybernetics,2009.
[4]J Lu,D Filev,K Prakah-Asante.From vehicle stability control to intelligent personal minder:real-time vehicle handling limit warning and driver style characterization[C].Computational Intelligence in Vehicles and Vehicular Systems,2009.
[5]MCG Quintero,JO Lopez,ACC Pinilla.Driver behavior classification model based on an intelligent driving diagnosis system[C].2012 15th International IEEE Conference on Intelligent Transportation Systems,2012.
[6]Pentland A,Liu A.Modeling and prediction of human behavior[J].Neural Computation,2009(11):229-242.
[7]A Wahab,TG Wen,N Kamaruddin.Understanding driver behavior using multi-dimensional CMAC[C].6th International Conference on Information,Communications&Signal Processing,2007.
[8]王建强,迟瑞娟,张磊.适应驾驶员特性的汽车追尾报警-避撞算法研究[J].公路交通科技,2009,1(10):22-28.(Wang Jian-qiang,Chi Rui-juan,Zhang Lei.Study on forward collision warning avoidance algorithm based on driver characteristics adaptation[J].Journal of Highway and Transportation Research and Development,2009,1(10):22-28.)
[9]Wang J,Lu M,Li K.Characterization of longitudinal driving behavior by measurable parameters[J].Transportation Research Record:Journal of the Transportation Research Board,2010,21(1):15-23
[10]吕岸,胡振程,陈慧.基于高斯混合隐马尔可夫模型的高速公路超车行为辨识与分析[J].汽车工程,2010,32(7):630-634.(Lv An,Hu Zhen-cheng,Chen Hui.Recognition on analysis on highway overtaking behavior based on gaussian mixture-hidden markov model[J].Automotive Engineering,2010,32(7):630-634)
[11]吴坚,姚琳琳,朱冰.基于高斯混合模型的驾驶员个人特质辨识[J].吉林大学学报:工学版,2015,45(1):38-43.(Wu Jian,Yao Lin-lin,Zhu Bing.Identification of driver individualities using gaussian mixture model[J].Journal of Jilin University:Engineering and Technology Edition,2015,45(1):38-43.)
[12]张磊,王建强,杨馥瑞.驾驶员行为模式的因子分析和模糊聚类[J].交通运输工程学报,2009,9(5):121-126.(Zhang Lei,Wang Jian-qiang,Yang Fu-rui.Factor analysis and fuzzy clustering of driver behavior pattern[J].Journal of Traffic and Transportation Engineering,2009,9(5):121-126.)