山东理工大学学报(自然科学版)
山東理工大學學報(自然科學版)
산동리공대학학보(자연과학판)
JOURNAL OF SHANDONG UNIVERSITY OF TECHNOLOGY(SCIENCE AND TECHNOLOGY)
2013年
4期
1-4
,共4页
商强%谭德荣%张晓琳%董春迎
商彊%譚德榮%張曉琳%董春迎
상강%담덕영%장효림%동춘영
驾驶行为%神经网络%驾驶倾向%交通安全
駕駛行為%神經網絡%駕駛傾嚮%交通安全
가사행위%신경망락%가사경향%교통안전
driving behavior%neural network%driver's tendency%traffic safety
驾驶倾向特征的实时辨识,对安全驾驶有积极的意义,但实时并准确识别驾驶倾向却是难点.通过分析驾驶倾向的外在表现,选取驾驶行为中的刹车频率、加油频率、刹车紧急程度和加油紧急程度作为评价指标,以实车试验获取的各评价指标数据为基础,运用BP神经网络构建驾驶倾向特征辨识模型,实现从驾驶行为到驾驶倾向的辨识.经验证,该模型识别率可达89.2%.
駕駛傾嚮特徵的實時辨識,對安全駕駛有積極的意義,但實時併準確識彆駕駛傾嚮卻是難點.通過分析駕駛傾嚮的外在錶現,選取駕駛行為中的剎車頻率、加油頻率、剎車緊急程度和加油緊急程度作為評價指標,以實車試驗穫取的各評價指標數據為基礎,運用BP神經網絡構建駕駛傾嚮特徵辨識模型,實現從駕駛行為到駕駛傾嚮的辨識.經驗證,該模型識彆率可達89.2%.
가사경향특정적실시변식,대안전가사유적겁적의의,단실시병준학식별가사경향각시난점.통과분석가사경향적외재표현,선취가사행위중적찰차빈솔、가유빈솔、찰차긴급정도화가유긴급정도작위평개지표,이실차시험획취적각평개지표수거위기출,운용BP신경망락구건가사경향특정변식모형,실현종가사행위도가사경향적변식.경험증,해모형식별솔가체89.2%.
Real-time recognition of driving tendency has a positive meaning in safe driving ,but it is difficult to identify driving tendency real time and accurately .We select the driver's braking fre-quency ,acceleration frequency ,emergency degree of braking and acceleration as the evaluation index through analyzing the external expressions of driving tendency ,obtain each evaluation in-dex data from the real vehicle test which will be used to construct training samples ,build driving tendency identification model and realize the recognition from the driving behavior to the driving tendency using BP neural network .This model is verified and the recognition rate is 89 .2% .