基于车联网BSM数据与路侧视频融合的港口集装箱卡车碰撞危险辨识方法

郭晓晗; 彭理群; 马定辉

doi:10.3963/j.jssn.1674-4861.2023.01.001

基于车联网BSM数据与路侧视频融合的港口集装箱卡车碰撞危险辨识方法

doi: 10.3963/j.jssn.1674-4861.2023.01.001

华东交通大学交通运输工程学院南昌 330013

基金项目:

国家重点研发计划项目 2020YFE0201200

国家自然科学基金项目 52062015

详细信息

作者简介:
郭晓晗（1996—），硕士研究生. 研究方向：智能交通系统. E-mail：zhuang99799717@163.com

通讯作者:
彭理群（1984—），博士，副教授. 研究方向：交通信息化与控制. E-mail：lq.peng@ecjtu.edu.cn

中图分类号: U491
计量
- 文章访问数: 990
- HTML全文浏览量: 333
- PDF下载量: 79
- 被引次数: 0
出版历程
- 收稿日期: 2022-01-24
- 网络出版日期: 2023-05-13

A Method of Identifying Collision Risk of Container Trucks in Port Terminal Areas under an Integrated Connected Vehicle BSM and Roadside Video Surveillance Data

School of Transportation Engineering, East China Jiaotong University, Nanchang 330013, China

摘要

摘要: 大型港口集装箱码头运输车辆调度频繁，堆场过道和交换区等区域视距狭窄，容易导致港口集装箱卡车与设施、作业人员和车辆发生擦碰事故。为提高智能集装箱卡车在港口密集区域的轨迹跟踪精度和行车安全感知能力，提出了一种车联网条件下融合车载终端基本安全消息（Basic Safety Messages，BSM）数据和路侧视频数据的集装箱卡车碰撞风险辨识方法。采用YOLOv5s算法提取视频监控范围内的目标车辆和作业人员，根据目标集卡大尺寸特点设计非极大值抑制锚框来提高目标识别准确度。运用透视变换原理将目标像素坐标转换成地理坐标，并应用Deep-SORT算法匹配每帧图像的车辆轨迹信息。应用交互式多模型方法（interactive multi-model，IMM）融合视频轨迹信息和车载单元（on-board units，OBU）定位数据，减小了目标机动过程中的观测误差。基于集卡融合轨迹结果，提出了1种新型的轨迹冲突风险评估模型，能够根据目标集卡与周围目标轨迹的相对运动状态实时感知车辆碰撞危险，该碰撞危险检测结果在实际场景中可通过路侧设备对车载终端和作业人员终端实时播发预警信息。针对集卡跟踪误差的实验结果表明：IMM自适应跟踪轨迹的平均均方根误差为0.29 m，比集卡自主跟踪轨迹误差提升81.05%；融合路侧监控视频与车载终端定位数据能够克服车辆自主定位系统在密集堆场环境下的误差增大问题。集卡碰撞危险辨识的结果表明：车辆碰撞危险识别结果（预设ETTC阈值为2 s）的召回率、精确度和准确度相对集卡自主感知分别提升了7.39%，4.27%，2.50%，更准确地辨识出了视线遮挡情况下的轨迹冲突风险。
- 智能交通 /
- 自动化码头 /
- 集卡碰撞危险辨识 /
- 车辆跟踪 /
- 车联网
Abstract: Large container terminals involve high-frequency transportation activities, and limited visibility in areas of stack aisles and exchange zones may easily lead to crashes between port container trucks and facilities, operators, and other vehicles. To improve the trajectory tracking accuracy and driving safety perception ability of intelligent container trucks in the densely populated port areas, a method for identifying container truck collision risks by integrating Connected Vehicle Basic Safety Messages (BSM) and roadside video surveillance data is proposed. A YOLOv5s algorithm is used to extract target vehicles and operators within the video surveillance, and a non-maximum suppression anchor box is designed based on the large size characteristics of the target container to improve detection accuracy. A perspective transformation principle is used to convert the target pixel coordinates into geographical coordinates, and a Deep-SORT algorithm is applied to match the vehicle trajectory information of each frame image. An interactive multi-model method (IMM) is used to fuse video trajectory information and vehicle positioning data of on-board units (OBU), reducing observation errors during target maneuvering process. Based on the trajectory fusion results, a new trajectory conflict risk assessment model is proposed, which can monitor vehicle collision risks in real-time according to the relative motion state of the target container and surrounding target trajectories. The detection of the collision risk can be broadcasted in real-time to on-board terminals and operator terminals through roadside equipment under most of practical scenarios. Experimental results show that the Root Mean Square Error (RMSE) of the IMM adaptive tracking method is only 0.29 m, which is 81.05% lower than that of the on-board tracking trajectory. It verifies that fusing roadside surveillance video with vehicle BSM positioning data can overcome the problem of increased errors from the on-board positioning systems under the dense stack environments. Study results also show that the recall rate, precision, and accuracy of collision risk identification results (with a pre-set ETTC threshold of 2 s) is improved by 7.39%, 4.27%, and 2.50%, respectively. The results indicate that the proposed method can more accurately identify collision risks in cases of obstructed visibility, when compared to the previous methods only using on-board detection techniques.
- intelligent transportation systems /
- automated container terminal /
- container truck collision risk identification /
- vehicle tracking /
- vehicle-to-infrastructure (V2I) communication

HTML全文

图 1 港口运输车辆行驶安全监测系统

Figure 1. Container truck driving safety monitoring system

下载: 全尺寸图片幻灯片

图 2 透视变换

Figure 2. Perspective transformation

下载: 全尺寸图片幻灯片

图 3 集卡数据融合

Figure 3. Container truck data fusion

下载: 全尺寸图片幻灯片

图 4 集卡碰撞危险场景

Figure 4. Container truck crash risk scenario

下载: 全尺寸图片幻灯片

图 5 集卡碰撞危险区域

Figure 5. Container truck crash risk area

下载: 全尺寸图片幻灯片

图 6 集卡跟踪结果

Figure 6. Tracking result of container trucks

下载: 全尺寸图片幻灯片

图 7 实验设计

Figure 7. Experimental design

下载: 全尺寸图片幻灯片

图 8 集卡跟踪轨迹展示

Figure 8. Tracking trajectory demonstration of container truck

下载: 全尺寸图片幻灯片

图 9 目标集卡位置误差

Figure 9. Location error of target container truck

下载: 全尺寸图片幻灯片

图 10 模型概率曲线

Figure 10. The model probability curve

下载: 全尺寸图片幻灯片

图 11 集卡碰撞危险辨识

Figure 11. Crash risk identification of container trucks

下载: 全尺寸图片幻灯片

图 12 集卡碰撞危险辨识ROC曲线

Figure 12. The ROC curve of truck crash risk identification

下载: 全尺寸图片幻灯片

表 1 不同YOLO版本性能对比

Table 1. Performance comparison of different YOLO versions

模型	图像尺寸px×px/	数据集	平均精确度/%	帧/s	时间/（s/帧）
YOLOv1	448×448	VOC2012	53.4	45	0.02
YOLOv2	544×544	VOC2012	49.0	40	0.03
YOLOv3	608×608	VOC2012	57.9	20	0.05
YOLOv4	608×608	VOC2012	69.7	62	0.02
YOLOv5s	640×640	VOC2012	75.4	140	0.01

下载: 导出CSV

表 2 集卡位置均方根误差均值

Table 2. Average RMSE of container truck positions 单位: m

跟踪方式	X坐标位置RMSE	Y坐标位置RMSE	平均RMSE
集卡自主跟踪	1.59	1.47	1.53
IMM自适应跟踪	0.31	0.26	0.29

下载: 导出CSV

表 3 碰撞危险辨识结果

Table 3. The results of crash risk identification 单位: %

感知方式	召回率	精确度	准确度
集卡自主感知	73.25	66.34	88.07
IMM自适应感知	78.66	69.17	90.27

下载: 导出CSV

参考文献(24)

[1]	FLAH A, MAHMOUDI C. Design and analysis of a novel power management approach, applied on a connected vehicle as V2V, V2B/I, and V2N[J]. International Journal of Energy Research, 2019, 43(13): 6869-6889.
[2]	SUN S, HU J, LI J, et al. An INS-UWB based collision avoidance system for AGV[J]. Algorithms, 2019, 12(2): 40-50. doi: 10.3390/a12020040
[3]	WANG T, TONG C, XU B. AGV navigation analysis based on multi-sensor data fusion[J]. Multimedia Tools and Applications, 2020(79): 5109-5124.
[4]	谢永良, 尹建军, 余承超, 等. 轮式AGV沿葡萄园垄道行驶避障导航算法与模拟试验[J]. 农业机械学报, 2018, 49(7): 13-22. XIE Y L, YIN J J, YU C C, et al. Obstacle avoidance navigation algorithm and analog experiment for wheeled AGV running along vineyard road[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(7): 13-22. (in Chinese)
[5]	裴珮, 卫泽坤. 自动化集装箱码头AGV防撞技术[J]. 港口装卸, 2019(1): 16-17, 63. doi: 10.3963/j.issn.1000-8969.2019.01.005 PEI P, WEI Z K. AGV anti-collision technology of automated container terminal[J]. Port Operation, 2019(1): 16-17, 63. (in Chinese) doi: 10.3963/j.issn.1000-8969.2019.01.005
[6]	丁一, 袁浩, 方怀瑾, 等. 考虑冲突规避的自动化集装箱码头AGV优化调度方法[J]. 交通信息与安全, 2022, 40(3): 96-107. doi: 10.3963/j.jssn.1674-4861.2022.03.010 DING Y, YUAN H, FANG H J, et al. An optimal scheduling method of AGVs at automated container terminal considering conflict avoidance[J]. Journal of Transport Information and Safety, 2022, 40(3): 96-107. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2022.03.010
[7]	CAO X, ZHU M. Research on global optimization method for multiple AGV collision avoidance in hybrid path[J]. Optimal Control Applications and Methods, 2021, 42(4): 1064-1080. doi: 10.1002/oca.2716
[8]	YANG Q, LIAN Y, XIE W. Hierarchical planning for multiple AGVs in warehouse based on global vision[J]. Simulation Modelling Practice and Theory, 2020(104): 102124.
[9]	张素云, 杨勇生, 梁承姬, 等. 自动化码头多AGV路径冲突的优化控制研究[J]. 交通运输系统工程与信息, 2017, 17 (2): 83-89. ZHANG S Y, YANG Y S, LIANG C J, et al. Optimal control of multiple AGV path conflict in automated terminals[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(2): 83-89. (in Chinese)
[10]	杨雅洁, 苌道方, 余芳. 考虑AGV避碰的自动化码头多资源协同调度[J]. 计算机工程与应用, 2020, 56(6): 246-253. YANG Y J, CHANG D F, YU F. Multi - resource coordinated scheduling of automated terminals considering AGV collision avoidance[J]. Computer Engineering and Applications, 2020, 56(6): 246-253. (in Chinese)
[11]	AL-QASSAB H, PANG S, AL-QIZWINI M, et al. Active safety system for connected vehicles[J]. SAE International Journal of Connected and Automated Vehicles, 2019, 2(3): 191-200.
[12]	柳涵, 黄妙华. 基于路侧单元视觉辅助的远程驾驶主动安全预警[J]. 重庆理工大学学报(自然科学), 2021, 35(7): 37-44. LIU H, HUANG M H. Remote driving active safety warning based on roadside unit vision assistance[J]. Journal of Chongqing University of Technology (Natural Science), 2021, 35(7): 37-44. (in Chinese)
[13]	李祎承, 胡钊政, 胡月志, 等. 基于GPS与图像融合的智能车辆高精度定位算法[J]. 交通运输系统工程与信息, 2017, 17(3): 112-119. LI Y C, HU Z Z, HU Y Z, et al. Accurate localization based on GPS and image fusion for intelligent vehicles[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(3): 112-119(. in Chinese)
[14]	ZHANG K J, WANG C, YU X Y, et al. Research on mine vehicle tracking and detection technology based on YOLOv5[J]. Systems Science & Control Engineering, 2022, 10 (1): 347-366.
[15]	CEBE M, ERDIN E, AKKAYA K, et al. Block4forensic: An integrated lightweight blockchain framework for forensics applications of connected vehicles[J]. IEEE communications magazine, 2018, 56(10): 50-57.
[16]	LI Y, WU D, LEE J, et al. Analysis of the transition condition of rear-end collisions using time-to-collision index and vehicle trajectory data[J]. Accident Analysis & Prevention, 2020(144): 105676.
[17]	MUTHALAGU R, BOLIMERA A, KALAICHELVI V. Lane detection technique based on perspective transformation and histogram analysis for self-driving cars[J]. Computers & Electrical Engineering, 2020(85): 106653.
[18]	ISLAM M, RAHMAN M, CHOWDHURY M, et al. Vision-based personal safety messages(PSMs)generation for connected vehicles[J]. IEEE Transactions on Vehicular Technology, 2020, 69(9): 9402-9416.
[19]	ZENG S, WANG X, DUAN X, et al. Kernelized mahalanobis distance for fuzzy clustering[J]. IEEE Transactions on Fuzzy Systems, 2020, 29(10): 3103-3117.
[20]	PENG X, MURPHEY Y L, LIU R, et al. Driving maneuver early detection via sequence learning from vehicle signals and video images[J]. Pattern Recognition, 2020(103): 107276.
[21]	BENTO L C, BONNIFAIT P, NUNES U J. Cooperative GNSS positioning aided by road-features measurements[J]. Transportation Research Part C: Emerging Technologies, 2017(79): 42-57.
[22]	ZHAO X, JING S, HUI F, et al. DSRC-based rear-end collision warning system-An error-component safety distance model and field test[J]. Transportation Research Part C: Emerging Technologies, 2019(107): 92-104.
[23]	XIE K, YANG D, OZBAY K, et al. Use of real-world connected vehicle data in identifying high-risk locations based on a new surrogate safety measure[J]. Accident Analysis & Prevention, 2019(125): 311-319.
[24]	谢济铭, 秦雅琴, 彭博, 等. 多车道交织区车辆跟驰行为风险判别与冲突预测[J]. 交通运输系统工程与信息, 2021, 21 (3): 131-139. XIE J M, QIN Y Q, PENG B, et al. Risk discrimination and conflict prediction of vehicle-following behavior in multi-lane weaving sections[J] Journal of Transportation Systems Engineering and Information Technology, 2021, 21(3): 131-139. (in Chinese)