基于双向RNN的私家车轨迹重构算法

doi:10.11959/j.issn.1000-436X.2020227

通信学报 ›› 2020, Vol. 41 ›› Issue (12): 171-181.doi: 10.11959/j.issn.1000-436X.2020227

基于双向RNN的私家车轨迹重构算法

肖竹¹, 钱鑫¹, 蒋洪波¹, 蔡成林², 曾凡仔¹

¹ 湖南大学信息科学与工程学院，湖南长沙 410082
² 湘潭大学信息工程学院，湖南湘潭 411105

修回日期:2020-10-24 出版日期:2020-12-25 发布日期:2020-12-01
作者简介:肖竹（1981- ），男，湖南涟源人，博士，湖南大学副教授、博士生导师，主要研究方向为下一代无线通信网、移动计算、测距与定位导航技术、大数据挖掘等。
钱鑫（1994- ），男，安徽合肥人，湖南大学硕士生，主要研究方向为车联网、测距与定位导航技术。
蒋洪波（1976- ），男，湖南长沙人，博士，湖南大学教授、博士生导师，主要研究方向为下一代无线通信网、移动计算。
蔡成林（1969- ），男，湖南双峰人，博士，湘潭大学教授、博士生导师，主要研究方向为GNSS定位与导航技术。
曾凡仔（1971- ），男，湖南郴州人，博士，湖南大学教授、博士生导师，主要研究方向为无线通信和车联网。
基金资助:
国家自然科学基金资助项目(U20A20181);国家自然科学基金资助项目(61732017);湖南省重点研发基金资助项目(2018GK2014)

Bidirectional RNN-based private car trajectory reconstruction algorithm

Zhu XIAO¹, Xin QIAN¹, Hongbo JIANG¹, Chenglin CAI², Fanzi ZENG¹

¹ College of Computer Science and Electronics Engineering, Hunan University, Changsha 410082, China
² College of Information Engineering, Xiangtan University, Xiangtan 411105, China

Revised:2020-10-24 Online:2020-12-25 Published:2020-12-01
Supported by:
The National Natural Science Foundation of China(U20A20181);The National Natural Science Foundation of China(61732017);The Key Research and Development Project of Hunan Province(2018GK2014)

摘要/Abstract

摘要：

在复杂的城市环境中，由于存在难以避免的GNSS定位信号中断现象以及车辆行驶过程中的误差累积，易造成所收集的车辆轨迹数据不准确和不完备，因此提出一种基于双向 RNN 的私家车轨迹重构算法，使用了GNSS-OBD轨迹采集设备收集车辆轨迹信息，利用多源数据融合实现双向加权轨迹重构。同时，在轨迹重构模型中引入神经算术逻辑单元（NALU），加强深度网络的外推能力并确保轨迹预测的精度，提高算法在应对城市复杂路段时轨迹重构的稳健性；选取了实际城市路段进行了测试实验，并和现有算法进行了对比分析。通过均方根误差（RMSE）以及Google Earth对重构轨迹进行可视化展示，实验结果验证了所提算法的有效性和可靠性。

关键词: 私家车, 车辆定位, 轨迹重构, 循环神经网络

Abstract:

To address the problem that in the complex urban environment, due to the inevitable interruption of GNSS positioning signal and the accumulation of errors during vehicle driving, the collected vehicle trajectory data was likely to be inaccurate and incomplete.a bidirectional weighted trajectory reconstruction algorithm was proposed based on RNN neural network.The GNSS-OBD trajectory acquisition device was used to collect vehicle trajectory information, and multi-source data fusion was adopted to achieve bidirectional weighted trajectory reconstruction.Furthermore, the neural arithmetic logic unit (NALU) was leveraged with the purpose of enhancing the extrapolation ability of deep network and ensuring the accuracy of trajectory reconstruction.For the evaluation, real-world experiments were conducted to evaluate the performance of the proposed method in comparison with existing methods.The root mean square error (RMSE) indicator shows the algorithm accuracy and the reconstructed trajectory is visually displayed through Google Earth.Experimental results validate the effectiveness and reliability of the proposed algorithm.

Key words: private car, vehicle positioning, trajectory reconstruction, RNN

中图分类号:

TN914

肖竹, 钱鑫, 蒋洪波, 蔡成林, 曾凡仔. 基于双向RNN的私家车轨迹重构算法[J]. 通信学报, 2020, 41(12): 171-181.

Zhu XIAO, Xin QIAN, Hongbo JIANG, Chenglin CAI, Fanzi ZENG. Bidirectional RNN-based private car trajectory reconstruction algorithm[J]. Journal on Communications, 2020, 41(12): 171-181.

图/表 13

表1

图1

图2

图3

图4

图5

图6

表2

图7

图8

图9

图10

表3

各方法重构轨迹的均方根误差"

算法	失效1（40 s）/m	失效2（40 s）/m	失效3（30 s）/m	失效4（66 s）/m
Bi-RNN-NALU	$3 . 76$	$6 . 625$	$4 . 627$	$12 . 939$
DR-OBD	55.031	44.815	48.609	39.793
DR-RNN	13.539	13.458	14.847	24.567
SVR	23.276	33.815	42.13	43.55
GPR	22.897	32.916	45.642	60.174

表3

参考文献 19

[1]	中国统计局. 中国统计黄页[M]. 北京: 中国统计出版社, 2018.
	National Bureau of Statistics. China statistical yearbook[M]. Beijing: China Statistical Publishing House, 2018.
[2]	YAN L , SHEN H Y , CHEN K . MobiT:distributed and congestion-resilient trajectory-based routing for vehicular delay tolerant networks[J]. IEEE/ACM Transactions on Networking, 2018,26(3): 1078-1091.
[3]	SHAN X N , HAO P , CHEN X H ,et al. Vehicle energy/emissions estimation based on vehicle trajectory reconstruction using sparse mobile sensor data[J]. IEEE Transactions on Intelligent Transportation Systems, 2018,20(2): 716-726.
[4]	CHIANG K W , TSAI G J , CHANG H W ,et al. Seamless navigation and mapping using an INS/GNSS/grid-based slam semi-tightly coupled integration scheme[J]. Information Fusion, 2019,50(1): 181-196.
[5]	BARRIOS C , MOTAI Y , HUSTON D . Trajectory estimations using smartphones[J]. IEEE Transactions on Industrial Electronics, 2015,62(12): 7901-7910.
[6]	BARRIOS C , MOTAI Y . Predicting vehicle trajectory[M]. Boca Raton: CRC Press, 2017.
[7]	HUANG Y R , XIAO Z , WANG D ,et al. Exploring individual travel patterns across private car trajectory data[J]. IEEE Transactions on Intelligent Transportation Systems, 2019,1(1): 1-15.
[8]	WANG D , FAN J J , XIAO Z ,et al. Stop-and-wait:discover aggregation effect based on private car trajectory data[J]. IEEE Transactions on Intelligent Transportation Systems, 2019,20(10): 3623-3633.
[9]	SUHR J K , JANG J , MIN D ,et al. Sensor fusion-based low-cost vehicle localization system for complex urban environments[J]. IEEE Transactions on Intelligent Transportation Systems, 2016,18(5): 1078-1086.
[10]	LIU L , . Enabling large-scale road data acquisition with an easy-to-install and sustainable setup[C]// Workshop on Mobisys Phd Forum. New York:ACM Press, 2017: 3-4.
[11]	LIU R , YUEN C , DO T N ,et al. Cooperative positioning for emergency responders using self IMU and peer-to-peer radios measurements[J]. Information Fusion, 2020,56(1): 93-102.
[12]	CARLOS M , RUIZ-GONZALEZ R , GOMEZ-GIL J . A data fusion system of GNSS data and on-vehicle sensors data for improving car positioning precision in urban environments[J]. Expert Systems with Applications, 2017,80(1): 28-38.
[13]	李璐璐, 华佳烽, 万盛 ,等. 基于高效信息缓存的位置隐私保护方案[J]. 通信学报, 2017,38(6): 148-157.
	LI L L , HUA J F , WAN S ,et al. Achieving efficient location privacy protection based on cache[J]. Journal on Communications, 2017,38(6): 148-157.
[14]	陶桦, 冯富琴, 肖鹏 ,等. 基于运行轨迹特征分析的车辆自组织网路由算法[J]. 通信学报, 2016,37(6): 144-153.
	TAO H , FENG F Q , XIAO P ,et al. Routing algorithm based on characteristics analysis of vehicle trace in vehicular ad hoc network[J]. Journal on Communications, 2016,37(6): 144-153.
[15]	YANG Z L , GUO X Q , CHEN Z M ,et al. RNN-Stega:linguistic steganography based on recurrent neural networks[J]. IEEE Transactions on Information Forensics and Security, 2019,14(5): 1280-1295.
[16]	TRASK A , HILL F , REED S ,et al. Neural arithmetic logic units[C]// The Thirty-second Annual Conference on Neural Information Processing Systems. Saarland:DBLP, 2018: 8046-8055.
[17]	WU Y , ZHU H , DU Q ,et al. A survey of the research status of pedestrian dead reckoning systems based on inertial sensors[J]. International Journal of Automation and Computing, 2019,16(1): 65-83.
[18]	XIAO Z , LI P , HAVYARIMANA V ,et al. GOI:a novel design for vehicle positioning and trajectory prediction under urban environments[J]. IEEE Sensors Journal, 2018,18(13): 5586-5594.
[19]	ZHANG W , LI Y , XIPNG W ,et al. Adaptive soft sensor for online prediction based on enhanced moving window GPR[C]// 2015 International Conference on Control,Automation and Information Sciences. Piscataway:IEEE Press, 2015: 291-296.

车辆ID	位置（经纬度）	GNSS信号状态	时间	加速度/(m.s^-2)	角速度/(rad.s^-1) (速度/m.s^-1)	方向(以正东方向逆时针为参考)	里程/km	发动机转数/(round.min^-1)	油门踏板位置
382704	东经：114°2′ 50′′北纬：22°35′ 40′′	强（9）	2018–07–0509:08:51	0.103 782	0.042 72518	150°	41 990.946	951	25.32
382704	东经：114°2′ 55′′北纬：22°35′ 32′′	强（7）	2018–07–0509:10:10	-0.389 651	-0.009 1614	120°	41 992.219	927	23.14
382704	东经：114°3′ 28′′北纬：22°35′ 40′′	弱（1）	2018–07–0509:14:55	0.443 45	-0.061 0311	100°	4 1994.083	885	19.23

测试阶段	GNSS信号失效时长/s	场景
测试1	40	交叉路口转弯
测试2	40	直线路段
测试3	30	直角转弯
测试4	66	立交桥路段

基于双向RNN的私家车轨迹重构算法

Bidirectional RNN-based private car trajectory reconstruction algorithm

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 19

相关文章 6

Metrics

推荐阅读 0

[1]	周炳朋, 马珊珊. 面向6G毫米波通信感知一体化的机动目标联合定位与测速[J]. 通信学报, 2023, 44(3): 81-92.
[2]	丁恩杰,刘忠育,刘亚峰,郁万里. 基于多维度和多模态信息的视频描述方法[J]. 通信学报, 2020, 41(2): 36-43.
[3]	石磊,杜军平,梁美玉. 基于RNN和主题模型的社交网络突发话题发现[J]. 通信学报, 2018, 39(4): 189-198.
[4]	李幼军,黄佳进,王海渊,钟宁. 基于SAE和LSTM RNN的多模态生理信号融合和情感识别研究[J]. 通信学报, 2017, 38(12): 109-120.
[5]	邵震洪,李文峰,吴怡,杨琼,沈连丰. 基于改进当前统计模型的模糊自适应车辆定位算法[J]. 通信学报, 2013, 34(7): 184-190.
[6]	邵震洪1,2，李文峰3，吴怡1，杨琼1，沈连丰1. 基于改进当前统计模型的模糊自适应车辆定位算法[J]. 通信学报, 2013, 34(7): 21-190.