基于双层隐马尔可夫模型的重型车辆行驶状态辨识方法研究

doi:10.3969/j.issn.1000-1093.2015.10.002

兵工学报 ›› 2015, Vol. 36 ›› Issue (10): 1832-1840.doi: 10.3969/j.issn.1000-1093.2015.10.002

基于双层隐马尔可夫模型的重型车辆行驶状态辨识方法研究

朱天军¹，孔现伟²，李彬³

(1.河北工程大学装备制造学院，河北邯郸 056038； 2.中国汽车技术研究中心汽车工程研究院，天津 300300；

收稿日期:2015-03-03 修回日期:2015-03-03 上线日期:2015-12-11
通讯作者: 朱天军 E-mail:happy.adam2012@gmail.com
作者简介:朱天军（1977—），男，副教授，硕士生导师
基金资助:
国家自然科学基金项目(51205105)；河北省自然科学基金项目(E2012402014)

Research on Driving Status Recognition of Heavy Duty Vehicle Based on Double-layer HMM Model

ZHU Tian-jun¹，KONG Xian-wei²，LI Bin³

(1.Equipment Manufacturing College, Hebei University of Engineering, Handan 056038,Hebei, China;2.Institute of Automotive Engineering, China Automotive Technology and Research Center, Tianjing 300300, China;3.CONCAVE Research Center, Department of Mechanical and Industrial Engineering, Concordia University, Montreal H3G 2W1,Canada)

Received:2015-03-03 Revised:2015-03-03 Online:2015-12-11
Contact: ZHU Tian-jun E-mail:happy.adam2012@gmail.com

摘要/Abstract

摘要： 为实时监测并发现重型车辆危险行驶姿态，在重型车辆侧翻预警系统中，采用一种基于双层隐马尔可夫模型的重型车辆行驶状态辨识方法，动态辨识重型车辆行驶状态。选取典型重型车辆行驶工况，采集相应工况数据，采用T-G检验法对数据序列进行剔除异常数据预处理。利用K-means算法设定重型车辆行驶状态的阈值。利用Baum-Welch算法对双层隐马尔可夫模型进行优化，应用优化后的隐马尔可夫模型进行重型车辆行驶状态在线辨识。辨识结果表明：该模型可准确、高效地辨识各个单一和复合工况下的车辆行驶状态，并且具有很好的实时性。

关键词: 兵器科学与技术, 双层隐马尔可夫模型, 重型车辆, 行驶状态辨识, 侧翻预警

Abstract: In order to detect the risky driving status of heavy duty vehicle, the double-layer hidden Markov model（HMM）is used for the rollover warning system of heavy duty vehicle, which can identify the driving status dynamically. The data is collected during the typical driving conditions, and the T-G test method is used to delete the abnormal data. The driving limit value of vehicle status is set based on K-means algorithm. Then finally Baum-Welch algorithm is used to optimize the proposed double-layer HMM, and the optimized HMM is used to identify the driving status in real-time. The identification results show that the proposed model can identify the driving status under every single driving condition as well as under the multiple driving conditions accurately and effectively. At the same time this model has a good real-time capacity.

Key words: ordnance science and technology, double-layer hidden Markov model, heavy duty vehicle, driving status identification, rollover warning

中图分类号:

U462.2

朱天军，孔现伟，李彬. 基于双层隐马尔可夫模型的重型车辆行驶状态辨识方法研究[J]. 兵工学报, 2015, 36(10): 1832-1840.

ZHU Tian-jun，KONG Xian-wei，LI Bin. Research on Driving Status Recognition of Heavy Duty Vehicle Based on Double-layer HMM Model[J]. Acta Armamentarii, 2015, 36(10): 1832-1840.

参考文献

［1］朱天军，宗长富，吴炳胜，等. 基于改进TTR算法的重型车辆侧翻预警系统[J].机械工程学报，2011,47(10): 88-94.
ZHU Tian-jun, ZONG Chang-fu, WU Bing-sheng, et al. Rollover warning system of heavy duty vehicle based on improved TTR algorithm [J]. Journal of Mechanical Engineering, 2011, 47(10): 88- 94.（in Chinese）
[2] Zong C F, Zhu T J, Wang C, et al. Research on multi-objective stability control algorithm of heavy tractor semi-trailer based on differential braking[J]. Chinese Journal of Mechanical Engineering, 2012, 25(1)：88-97.
[3] Lin C C, Wang M S. Vehicle speeding early warning model using frame feature detection and HMM [C]∥IEEE 15th International Symposium on Consumer Electronics. Singapore：IEEE， 2011:241- 244.
[4] Maghsood R, Johannesson P. Detection of the curves based on lateral acceleration using hidden Markov models [J]. Procedia Engineering, 2013, 66：425-434.
[5] Gadepally V, Kurt A, Krishnamurthy A. Driver/vehicle state estimation and detection[C]∥14th International IEEE Conference on Intelligent Transportation Systems. Washington DC：IEEE，2011：582-587.
[6] Pentland A P . Graphical models for driver behavior recognition in a smart car[C]∥Proceedings of the IEEE Intelligent Vehicles Symposium 2000.Dearborn,MI, US：IEEE，2000:7-12.
[7] 宗长富,杨肖,王畅，等.汽车转向时驾驶员驾驶意图辨识与行为预测[J].吉林大学学报:工学版，2009, 39(1)： 28-32.
ZONG Chang-fu, YANG Xiao, WANG Chang，et al. Driving intentions identification and behaviors prediction in car lane change[J]. Journal of Jilin University :Engineering and Technology Edition, 2009, 39(1)：28-32.（in Chinese）
[8] 吕岸,胡振程,陈慧.基于高斯混合隐马尔科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.
LYU An, HU Zhen-cheng, CHEN Hui. Recognition and analysis on highway overtaking behavior based on gaussia mixture-hidden Markov model [J]. Journal of Automotive Engineering, 2010, 32(7)： 630-634.（in Chinese）
[9] Rabiner L R. A tutorial on hidden Markov models and selected applications in speech recognition[J]. Proceedings of the IEEE, 1989，77(2):257-286.
[10] Yu S Z, Hisashi K. An efficient forward-backward algorithm for an explicit-duration hidden markove model [J]. IEEE Signal Processing Letters, 2003, 10(1): 11-14.
[11] 蔡天净,唐濠. Savitay-Golay平滑滤波器的最小二乘拟合原理综述[J].数字通信，2011（1）：63-68.
CAI Tian-jing, TANG Hao. The principle of least-squares review of Savitay-Golay smoothing filter [J]. Digital Telecommunications, 2011(1)：63-68.（in Chinese）
[12] 肖树臣，秦玉勋，韩吉庆.基于格拉布斯法的试验数据分析方法[J]. 弹前与制导学报，2007,27(1):275-277.
XIAO Shu-chen, QING Yu-xun, HAN Ji-qing. A test analysis method based on Grubbs in experiments [J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2007, 27(1):275-277.（in Chinese）
[13] 全国文献工作标准化技术委员会.GB4883—85 数据的统计处理和解释正态样本异常值的判定和处理[S].北京：中国标准出版社，1985.
Technical Committee of National Literature Work Standardization. GB4883—85 The statistical processing and interpretation of data processing and decision of the abnormal value of normal sample [S]. Beijing: China Standard Press,1985.（in Chinese）
[14] Agha M E L. Efficient and fast initialization algorithm for K-means clustering [J]. International Journal Intelligent Systems and Applications, 2012(1): 21-31.

[1]	王玮琪，王伟达，孙晓霞，张渊博，刘城，项昌乐. 重型车辆多挡并联混合动力系统的联合优化控制策略[J]. 兵工学报, 2021, 42(10): 2242-2250.
[2]	杨绍卿, 翟振东, 王宪成，王雪. 重型车辆柴油机气缸当量漏气面积检测模型研究[J]. 兵工学报, 2018, 39(10): 1883-1891.
[3]	娄伟鹏，郑长松，李和言，陈建文，张周立，马源. 高速《Symbol》v湿式换挡离合器排油阀临界开启和关闭压力研究[J]. 兵工学报, 2017, 38(9): 1665-1672.
[4]	王宪成，杨绍卿，马宁，赵文柱. 车辆柴油机缸套动载荷磨损计算模型研究[J]. 兵工学报, 2017, 38(9): 1673-1680.
[5]	孙皓泽，常天庆，王全东，孔德鹏，戴文君. 一种基于分层多尺度卷积特征提取的坦克装甲目标图像检测方法[J]. 兵工学报, 2017, 38(9): 1681-1691.
[6]	张玉燕，孙莎莎，王振春，曹海要，战再吉. 高速载流电枢表面瞬态温度场仿真与测量[J]. 兵工学报, 2017, 38(9): 1692-1698.
[7]	李臣明，宦超，石怀龙. 某箱式火箭炮对面目标分布式杀伤最优火力分配[J]. 兵工学报, 2017, 38(9): 1699-1704.
[8]	雷娟棉，张嘉炜，谭朝明. 小攻角下船尾外形对旋转弹丸马格努斯效应影响的数值研究[J]. 兵工学报, 2017, 38(9): 1705-1715.
[9]	李泽，闫晓鹏，栗苹，郝新红，王建涛. 扫频式干扰对调频多普勒引信的干扰机理研究[J]. 兵工学报, 2017, 38(9): 1716-1722.
[10]	张浩为，谢军伟，张昭建，宗彬锋，陈唐军. 基于混合自适应遗传算法的相控阵雷达任务调度[J]. 兵工学报, 2017, 38(9): 1761-1770.
[11]	李茂，侯海量，朱锡，黄晓明，李典，陈长海，胡年明. 结构间隙对芳纶纤维增强复合装甲结构抗侵彻性能的影响[J]. 兵工学报, 2017, 38(9): 1797-1805.
[12]	王鉴, 韩焱, 王黎明, 张丕状, 陈平. 弹丸在膛内运动的回波信号瞬时频率估计方法研究[J]. 兵工学报, 2017, 38(9): 1806-1814.
[13]	刘海鸥，张文胜，徐宜，赵梓烨. 基于核密度估计的履带车辆传动轴载荷谱编制[J]. 兵工学报, 2017, 38(9): 1830-1838.
[14]	姜缪文，闫健卓，陈继民. 基于拓扑优化技术的军用头盔内胆结构三维打印[J]. 兵工学报, 2017, 38(9): 1845-1853.
[15]	张磊，李世民，朱刚. 基于时间连续灰色Markov模型的维修器材需求预测方法研究[J]. 兵工学报, 2017, 38(9): 1862-1866.

基于双层隐马尔可夫模型的重型车辆行驶状态辨识方法研究

Research on Driving Status Recognition of Heavy Duty Vehicle Based on Double-layer HMM Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价