《基于方向盘握力的疲劳驾驶检测研究_沙春发.docx》由会员分享，可在线阅读，更多相关《基于方向盘握力的疲劳驾驶检测研究_沙春发.docx（6页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、基于方向盘握力的疲劳驾驶检测研究 沙春发李瑞张明明 (江苏大学图形技术研宄所，镇江 212013) 摘要选取 6 位普通驾驶员参加实际道路驾驶实验，在实验中同步检测驾驶员对方向盘的握力信号和驾驶员的脑电信号。 通过独立样本 T 检验，从时域和时频域两方面筛选出 5 个与疲劳驾驶密切相关的握力信号特征，将它们作为输入信息；提取脑 电功率谱比值作为衡量疲劳驾驶的信号特征，将其作为输出信息 ;通过 BP 神经网络方法，以输入信息和输出信息建立基于方 向盘握力的疲劳驾驶检测模型。使用部分驾驶数据对检测模型进行验证，结果显示此数学模型对疲劳驾驶的识别率达到 87. 0%，说明方向盘握力信号可作为检测疲劳
2、驾驶的有效数据。 关键词疲劳驾驶检测 方向盘握力 脑电 神经网络 中图法分类号 U491.254; 文献标志码 B 据不完全统计，疲劳驾驶导致数以万计的人员 死亡，超过 30%的交通事故与之有关 1。由于疲劳 驾驶的高发性与危险性，实时检测疲劳驾驶对疲劳 驾驶预警、预防驾驶事故有重大意义。目前，国内外 主要通过车辆、驾驶员等对象在行驶中的实时信息 来研宄如何检测驾驶员疲劳状态。其监测的信息主 要有三个类别： 基于驾驶员表情特性的检 测 24。通过检测驾驶员表情特征来判断驾驶员的 疲劳状态，如瞳孔直径变化、眼睑活动、眼睛闭合状 态、面部表情、头部移动等。该类检测方法为非接触 式测量，对驾驶员驾驶
3、不会造成影响，简单易行。但 是疲劳判断的准确性受驾驶姿态、信息采集方式和 环境等影响较大； 基于车辆行驶特征的 检 测 5_7。通过检测车辆行驶参数和驾驶员操作行为 来判断疲劳状态，如车道偏离、转向盘转角、转向盘 转动频率等。该类检测方法较为直接地反映了驾驶 质量。但实际应用时，该方法受道路环境、汽车型号 及个人驾驶习惯等影响较大 ; 基于驾驶员生理信 号特征的检测 8_1 0 prt = - j - (7) 式 （ 7)中，当 4 0 成立时， c; 0值为 1，否则值 为 0。 (8) 用 Z/mr,表示小波系数正系数和与负系数 绝对值的和之比的对数值，用其表征对应的时间尺 度下驾驶员握力
4、増加的量和减少的量之间的比值， 计算公式如式 （ 8) Q X 1% Ipnr； = lg - (8) X 11* Uij Sotelo M A, ei al. Real-time system for monitoring driver vigilance. Intelligent Transportation Systems IEEE Transactions on, 2004; 7(1) :63 77 4 Dasgupta A， George A, Happy S L et al. Avision 七 ased system for monitoring the loss of atte
5、ntion in automotive drivers. Transactions on Intelligent Transportation Systems 2013； 14(4) : 18251838 5 Eskandarian A， Mortazavi A. Evaluation of asmart algorithm for commercial vehicle driver drowsiness detection. IEEE Intelligent Vehicles Symposium，2007 :553559 6 Bertozzi M Bombini L Broggi A, et
6、 al. GOLD: a framework for developing intelligent-vehicle vision applications. IEEE Intelligent Systems, 2008; 23(1) :6971 7 白艳，管欣 .车辆横摆转动惯量及轮胎侧偏刚度识别方法 的研究 科学技术与工程， 2015; 15(9) :125128 Bai Y， Guan X. Research on system identification of vehicle moment of inertia about the Z axis and tire cornering s
7、tiffness. Science Technology & Engineering, 2015； 15 (9) ： 125128 8 Lai S K L， Craig A. A critical review of the psychophysiology of driver fatigue. Biological Psychology, 2001; 55(3) :173194 9 Lai S K L Craig A， Boord P et al. Development of an algorithm for an EEG-based driver fatigue countermeasu
8、re. Journal of Safety Research, 2003; 34(3) :321 328 10 Zuzewicz K. Heart rate variability (HRV) and muscular system activity (EMG) in cases of crash threat during simulated driving of a passenger car. International Journal of Occupational Medicine & Environmental Health, 2013; 26(5) :710723 11 Baro
9、nti F, Lenzi F Roncella R et al. Distributed sensor for steering wheel rip force measurement in driver fatigue detection. 2009 Design, Automation and Test in Europe. Maine: IEEE， 2009: 894897 12 于晓东 .基于驾驶人生理指标的驾驶疲劳量化方 法研宄 .长 春 :吉林大学，2015 Yu X D. Study on the quantitative method of driving fatigue ba
10、sed on the driver physiological indexes. Changchun: Jilin University, 2015 13 王兵 .脑电信号中伪迹去除的综合研宄 .杭州 ： 浙江大 学， 2010 Wang B. Comprehensive study on removal of artifacts from EEG data. Hangzhou： Zhejiang University 2010 14 房瑞雪，赵晓华，荣建，等 .基于脑电信号的驾驶疲劳研究 . 公路交通科技，2009; 26 (S1) : 124126 Fang R X? Zhao X H R
11、ong J et al. Study on driving fatigue based on EEG signals. Journal of Highway & Transportation Research & Development， 2009; 26 (SI) : 124126 15 Li F， Wang X W, Lu B L. Detection of driving fatigue based on grip force on steering wheel with wavelet transformation and support vector machine. Neural
12、Information Processing. Daegu: Springer Berlin Heidelberg, 2013: 141148 16 戴硕，罗海，黄河，等 .基 于 声 信 号 处 理 的 交 通 事 故 自 动检测算法 .电子技术， 2010; 37(10) : 4 6 Dai S Luo H， Huang H, et al. A traffic accident detection algorithm based on audio signal processing. Electronic Technology, 2010; 37(10) : 46 304 科学技术与工程
13、16 卷 Detecting Fatigue Driving Based on Steering Wheel Grip Force SHA Chuna? LI Rui, ZHANG Mingning (Insititute of Graphic Technology， Jiangsu University， Zhenjiang 212013， P. R. China) Abstract Six drivers were asked to driver on a real road. Then， the steering wheel rip force signal and the driver
14、s EEG signal were detected at the same time. From time domain and time frequency domain? five steering wheel rip force features were chosen based on t test. These features were used as input which were closely related to driving fatigue. The ratio of EEG power spectrum which could show the state of
15、fatigue driving was used as output. Based on BP neural network, a prediction model of fatigue driving based on steering wheel rip force was established with the input and the output. A part of the real road driving data was used to verify the model， the results show that the model has 87. 0% recogni
16、tion rate of fatigue driving. So， steering grip force signal data is effective for detecting fatigue driving. Key words driver fatigue detecting BP neural network steering wheel grip force EEG (上接第 298 页 ) Application of the New Flexible Equipment for Bridge Anti-collision in Mountain Bridges ZHANG
17、Xingdng1 , CHEN Ming-flong1 , CHEN Guo-yu2 CHEN Ming1 (Key Laboratory of Hydraulic and Waterway Engineering of the Ministry of Education, Chongqing Jiaotong University1， Chongqing 400074， P. R. China; Institute of Shanghai Ocean Steel Structure2， Shanghai 201204， P. R. China) Abstract The channel in
18、 mountain rivers which have many beaches and a big velocity is complex and narrow. Moreover， because of a large number of bridges， there is a huge possibility of ship4)ridge collision. The studies on Shuangbei bridge in Chongqing are expatiated and analyzed by ship4)ridge collision risk, ship4)ridge
19、 collision force and anti-collision projects， based on the existing anti-collision technology at home and abroad. By analyzing the projects comparison in detail， it is found that the new flexible equipment can reduce about 55% ship4)ridge collision force. At the same time， the bow is not only pushed
20、 away but also changed directions during the accident which can protect the bridges and ships effectively. What is more? the equipment can adapt to the reach with a big velocity， gradient and a high head. In a word， the new flexible equipment has a broad application prospects in mountain bridges. Key words mountain rivers bridges ship4)ridge collision risk ship4ridge collision force anti- collision equipment