混合动力公交车驾驶员驾驶特性辨识算法  

Identification Algorithm of Driving Behavior for Hybrid Bus Driver

作　　者：严英[1] 朱道伟 叶鹏[3] 杨钢 

机构地区：[1]天津职业技术师范大学,天津300222 [2]军事交通学院军用车辆系,天津300161 [3]军事交通学院科研部,天津300161

出　　处：《军事交通学院学报》2017年第2期44-47,共4页Journal of Military Transportation University

基　　金：国家自然科学基金资助项目(51307119);天津市局级预研项目(KRKC011503)

摘　　要：针对混合动力公交车驾驶员驾驶特性与标准控制参数不匹配而导致车辆续驶里程降低的问题,研究同一道路工况下的驾驶员驾驶行为特征参数描述方法以及驾驶员驾驶特性辨识算法。通过提取时域场中与能耗相关的特征变量,构建基于高斯混合模型(GMM)算法的驾驶员驾驶特性辨识模型。结果表明:踏板开度的倒频谱信号比时域信号更能反映驾驶员的驾驶倾向;基于特征变量的时频域信号,并结合线路运行特征优化模型参数,最终模型辨识精度高于93%。Considering the mismatch between driving behavior of hybrid bus driver and standard control parameter reduces driving range, the paper studies the description method of driving behavior characteristic parameter and identification algo- rithm of driving behavior in the same road condition. It establishes an identification model of driving behavior based on GMM （Gaussian mixture model） by extracting characteristic variable related to energy consumption in time domain. The re- sult shows that the cepstrum signal of pedal can reflect drivers＇ driving tendency rather than time domain signal, and the precision of the model can reach 93% by optimizing the model parameter according to the running characteristics based on time-frequency domain signal of characteristic variable.

关 键 词：混合动力公交车 驾驶特征 高斯混合模型 驾驶员辨识 

分 类 号：U469.72[机械工程—车辆工程]