Fusion method of road section average speed based on BP neural network and D-S evidence theory
-
摘要: 为精确估计路段平均速度, 提出了基于BP神经网络与D-S证据理论的路段平均速度融合方法。通过训练完成的BP神经网络估计概率密度函数值, 进而通过D-S证据理论进行数据融合, 整合了BP神经网络自学习的特点与D-S证据理论推理的能力。提出了融合方法的框架, 给出了具体的计算模型。利用京藏高速公路上的实测浮动车数据、微波检测器数据、车牌识别数据对融合方法进行了验证, 并分析了当微波检测器失效时融合方法的鲁棒性。分析结果表明: 融合数据的平均绝对误差百分率比仅使用浮动车数据或微波检测器数据分别提高了7.90%、20.72%, 融合方法能够得到较好的效果。微波检测器失效的情况下, 融合精度有所下降, 但融合数据的误差仍然小于仅使用浮动车数据的误差, 说明融合方法具有一定的鲁棒性。Abstract: In order to estimate road section average speed accurately, a fusion method of road section average speed based on BP neural network and D-S evidence theory was proposed.The values of probability density function were estimated by trained BP neural network, and the data were fused by D-S evidence theory.The self-learning ability of BP neural network and the reasoning ability of D-S evidence theory were combined in the fusion method.The framework and model of the fusion method were presented, and each process of the method was analyzed.The fusion method was verified by using floating car data (FCD), microwave detector data, and license plate recognition data from Beijing-Xizang Expressway.The robustness of the fusion method was verified in the case that microwave detector failed to work.Analysis result indicates that the mean absolute percentage errors of fusion data are 7.90%, 20.72% better than that of FCD and microwave detector data respectively.When microwave detector fail to work, the fusion accuracy reduces, but the errors of fusion data is still smaller than that of FCD, and the fusion method is proved to be robustness.
-
[1] HALL D L, LLINAS J. An introduction to multisensor data fusion[J]. Proceedings of the IEEE, 1997, 85 (1): 6-23. doi: 10.1109/5.554205 [2] 刘红红, 杨兆升. 基于数据融合技术的路段出行时间预测方法[J]. 交通运输工程学报, 2008, 8 (6): 88-92. doi: 10.3321/j.issn:1671-1637.2008.06.017LIU Hong-hong, YANG Zhao-sheng. Estimating methods of link travel times based on data fusion technology[J]. Journal of Traffic and Transportation Engineering, 2008, 8 (6): 88-92. (in Chinese). doi: 10.3321/j.issn:1671-1637.2008.06.017 [3] 邹亮, 徐建闽, 朱玲湘, 等. 基于浮动车移动检测与感应线圈融合技术的行程时间估计模型[J]. 公路交通科技, 2007, 24 (6): 114-117. doi: 10.3969/j.issn.1002-0268.2007.06.026ZOU Liang, XU Jian-min, ZHU Ling-xiang, et al. Estimation model of travel time based on fusion technique from probe vehicle and crossing data[J]. Journal of Highway and Transportation Research and Development, 2007, 24 (6): 114-117. (in Chinese). doi: 10.3969/j.issn.1002-0268.2007.06.026 [4] 何友, 彭应宁, 陆大. 多传感器数据融合模型综述[J]. 清华大学学报: 自然科学版, 1996, 36 (9): 14-20. doi: 10.3321/j.issn:1000-0054.1996.09.002HE You, PENG Ying-ning, LU Da-jin. Survey of multisensor data fusion models[J]. Journal of Tsinghua University: Science and Technology, 1996, 36 (9): 14-20. (in Chinese). doi: 10.3321/j.issn:1000-0054.1996.09.002 [5] 仲崇权, 张立勇, 杨素英, 等. 多传感器分组加权融合算法研究[J]. 大连理工大学学报, 2002, 42 (2): 242-245. doi: 10.3321/j.issn:1000-8608.2002.02.024ZHONG Chong-quan, ZHANG Li-yong, YANG Su-ying, et al. Study of grouping weighted fusion algorithm for multi-sensor[J]. Journal of Dalian University of Technology, 2002, 42 (2): 242-245. (in Chinese). doi: 10.3321/j.issn:1000-8608.2002.02.024 [6] LIU Hao, ZHANG Ke, WANG Zi-lei, et al. A comparison of existing algorithms for travel time estimation[C]//ASCE. Proceedings of 2009International Conference on Transportation Engineering. Chengdu: ASCE, 2009: 189-194. [7] 李慧兵, 杨晓光. 面向行程时间预测准确度评价的数据融合方法[J]. 同济大学学报: 自然科学版, 2013, 41 (1): 60-65. doi: 10.3969/j.issn.0253-374x.2013.01.010LI Hui-bing, YANG Xiao-guang. Data fusion method for accuracy evaluation of travel time forecast[J]. Journal of Tongji University: Natural Science, 2013, 41 (1): 60-65. (in Chinese). doi: 10.3969/j.issn.0253-374x.2013.01.010 [8] LI Hui-bing, YANG Xiao-guang, LIU Hao-de. Research on multi-source data fusion based on loop detector data and FCD (floating car data)[C]//YAN Xin-ping, YI Ping, WU Chaozhong, et al. 2011Multimodal Approach to Sustained Transportation System Development: Information, Technology, Implementation. Wuhan: ASCE, 2011: 495-501. [9] GUO Jian-hua, XIA Jing-xin, SMITH B L. Kalman filter approach to speed estimation using single loop detector measurements under congested conditions[J]. Journal of Transportation Engineering, 2009, 135 (12): 927-934. doi: 10.1061/(ASCE)TE.1943-5436.0000071 [10] BYON Y J, SHALABY A, ABDULHAI B, et al. Traffic data fusion using SCAAT Kalman filters[C]//TRB. Transportation Research Board 89th Annual Meeting. Washington DC: TRB, 2010: 1-16. [11] CHOI K, CHUNG Y S. A data fusion algorithm for estimating link travel time[J]. Journal of Intelligent Transportation Systems: Technology, Planning, and Operations, 2002, 7 (3/4): 235-260. [12] FAOUZI N E, CHARLINE S. Travel time estimation by evidential data fusion[J]. Recherche Transports Sécurité, 2000 (68): 15-30. [13] FAOUZI N E, LEFEVRE E. Classifiers and distance-based evidential fusion for road travel time estimation[C]//SPIE. 2006Defense and Security Symposium. Orlando: International Society for Optics and Photonics, 2006: 1-16. [14] FAOUZI N E. Data-driven aggregative schemes for multisource estimation fusion: a road travel time application[C]//SPIE. 2004 Defense and Security Symposium. Orlando: International Society for Optics and Photonics, 2004: 351-359. [15] 李嘉, 刘春华, 胡赛阳, 等. 基于交通数据融合技术的行程时间预测模型[J]. 湖南大学学报: 自然科学版, 2014, 41 (1): 33-38. doi: 10.3969/j.issn.1008-1763.2014.01.006LI Jia, LIU Chun-hua, HU Sai-yang, et al. A travel time prediction model based on traffic data fusion technology[J]. Journal of Hunan University: Natural Sciences, 2014, 41 (1): 33-38. (in Chinese). doi: 10.3969/j.issn.1008-1763.2014.01.006 [16] 李瑞敏, 陈熙怡. 多源数据融合的道路旅行时间估计方法研究[J]. 公路交通科技, 2014, 31 (2): 99-103. doi: 10.3969/j.issn.1002-0268.2014.02.017LI Rui-min, CHEN Xi-yi. Study on methods of travel time estimation based on multi-source data fusion[J]. Journal of Highway and Transportation Research and Development, 2014, 31 (2): 99-103. (in Chinese). doi: 10.3969/j.issn.1002-0268.2014.02.017 [17] BACHMANN C, ABDULHAI B, ROORDA M J, et al. A comparative assessment of multi-sensor data fusion techniques for freeway traffic speed estimation using microsimulation modeling[J]. Transportation Research Part C: Emerging Technologies, 2013, 26 (1): 33-48. [18] 徐从富, 耿卫东, 潘云鹤. 面向数据融合的DS方法综述[J]. 电子学报, 2001, 29 (3): 393-396. doi: 10.3321/j.issn:0372-2112.2001.03.027XU Cong-fu, GENG Wei-dong, PAN Yun-he. Review of Dempster-Shafer method for data fusion[J]. Acta Electronica Sinica, 2001, 29 (3): 393-396. (in Chinese). doi: 10.3321/j.issn:0372-2112.2001.03.027 [19] DEMPSTER A P. A generalization of Bayesian inference[J]. Journal of the Royal Statistical Society, 1968, 30 (2): 205-247. [20] KADALI B R, RATHI N, PERUMAL V. Evaluation of pedestrian mid-block road crossing behaviour using artificial neural network[J]. Journal of Traffic and Transportation Engineering: English Edition, 2014, 1 (2): 111-119. doi: 10.1016/S2095-7564(15)30095-7 [21] YUE H. Chaotic time series prediction for duffing system based on optimized BP neural network[J]. Information Technology Journal, 2013, 12 (19): 5401-5405. doi: 10.3923/itj.2013.5401.5405 [22] 刘桂莲, 王福林, 索瑞霞. BP神经网络算法的改进及其应用[J]. 农业系统科学与综合研究, 2010, 26 (2): 170-173. doi: 10.3969/j.issn.1001-0068.2010.02.009LIU Gui-lian, WANG Fu-lin, SUO Rui-xia. An improved method of BP neural network and its application[J]. System Sciences and Comprehensive Studies in Agriculture, 2010, 26 (2): 170-173. (in Chinese). doi: 10.3969/j.issn.1001-0068.2010.02.009 [23] 宋俪婧, 陈金川, 石建军, 等. 应用车辆牌照自动识别系统自动检测行程延误的算法研究[J]. 交通运输工程与信息学报, 2008, 6 (2): 107-112. https://www.cnki.com.cn/Article/CJFDTOTAL-JTGC200802021.htmSONG Li-jing, CHEN Jin-chuan, SHI Jian-jun, et al. Algorithm rsearch of auto-detecting the travel delay information with vehicle license plate automatic recognition[J]. Journal of Transportation Engineering and Information, 2008, 6 (2): 107-112. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JTGC200802021.htm [24] 何小荣, 陈丙珍, 赵晓光, 等. 改善BP网络检验效果的研究[J]. 清华大学学报: 自然科学版, 1995, 35 (3): 31-36. https://www.cnki.com.cn/Article/CJFDTOTAL-QHXB503.005.htmHE Xiao-rong, CHEN Bing-zhen, ZHAO Xiao-guang, et al. Study on improving testing results of BP neural networks[J]. Journal of Tsinghua University: Science and Technology, 1995, 35 (3): 31-36. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-QHXB503.005.htm [25] EBENEZER B, HARRIS E, NYABADZA F. Forecasting Buruli ulcer disease in Ashanti Region of Ghana using BoxJenkins approach[J]. American Journal of Mathematics and Statistics, 2013, 3 (3): 166-177. [26] 李月, 徐余法, 陈国初, 等. D-S证据理论在多传感器故障诊断中的改进及应用[J]. 东南大学学报: 自然科学版, 2011, 41 (增): 102-106. https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX2011S1021.htmLI Yue, XU Yu-fa, CHEN Guo-chu, et al. Improvement and application of D-S evidence theory in multi-sensor fault diagnosis system[J]. Journal of Southeast University: Natural Science Edition, 2011, 41 (S): 102-106. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX2011S1021.htm -
下载:
点击查看大图
图(6) / 表(2)
计量
- 文章访问数: 681
- HTML全文浏览量: 115
- PDF下载量: 900
- 被引次数: 0
