基于轨迹预测和极限梯度提升的驾驶意图识别

doi:10.13229/j.cnki.jdxbgxb.20230479

摘要/Abstract

摘要：

为实现智能网联车对周围车辆驾驶意图的准确辨识，提出了一种基于轨迹预测与极限梯度提升算法（XGBoost）的驾驶意图识别框架。首先，通过标注车辆历史轨迹的驾驶意图来建立离线训练数据集。其次，构建驾驶意图识别框架，通过混合示教的长短时记忆网络（LSTM）模块预测目标车辆的未来轨迹，XGBoost模块融合历史轨迹和未来轨迹来识别出驾驶意图。最后，采用实际道路数据集NGSIM（Next Generation SIMulation）US101和I-80路段来验证模型框架。实验结果表明：该方法在4 s历史轨迹预测3 s未来轨迹处识别准确率可达97.7%，表现出较强的驾驶意图识别能力。实现代码见网站：https：∥gitee.com/fanghz-colin/lstm-xgboost.git。

关键词: 交通工程, 长短时记忆网络, 极限梯度提升, 智能网联车, 驾驶意图识别

Abstract:

To achieve accurate identification of the driving intentions of surrounding vehicles by intelligent connected vehicles， a driving intention recognition framework based on trajectory prediction and extreme gradient boosting （XGBoost） algorithm is proposed. Firstly， we attach the driving intention label to the vehicle's historical trajectory sequence and build the offline training dataset. Then， a driving intention recognition framework is constructed. A mixed teacher force long short-term Memory （LSTM） module is used to predict the future trajectory. The XGBoost module concatenates the historical and future trajectory and recognizes driving intention （left lane change， lane keeping， and right lane change）. Finally， the model is verified on the real road datasets Next Generation SIMulation（NGSIM） US101 and I-80 sections. The experimental results show that the proposed model outperforms the other methods in metrics precision， recall， F1 score， and accuracy. The recognition accuracy can reach 97.7% in the prediction of 4 s historical trajectory and 3 s future trajectory， which shows good performance in driving intention recognition. The code can be obtained at：https：∥gitee.com/fanghz-colin/lstm-xgboost.git.

Key words: traffic engineering, long short-term memory, extreme gradient boosting, intelligent connected vehicle, driving intention recognition

中图分类号:

U495

方华珍,刘立,顾青,肖小凤,孟宇. 基于轨迹预测和极限梯度提升的驾驶意图识别[J]. 吉林大学学报(工学版), 2025, 55(2): 623-630.

Hua-zhen FANG,Li LIU,Qing GU,Xiao-feng XIAO,Yu MENG. Driving intention recognition based on trajectory prediction and extreme gradient boosting[J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 623-630.

图/表 13

图1

表1

表2

模型输入特征描述"

组别	参数	特征描述	数量
TVF	$x t, y t$	目标车辆的横向和纵向位移	2
	$v x t, v y t$	目标车辆的横向和纵向速度，通过相邻轨迹点之间的位移可得： $v x t = 10 x t + 1 - x t, v y t = 10 y t + 1 - y t$	2
	$a x t, a y t$	目标车辆的横向和纵向加速度，通过相邻轨迹点之间的速度可得： $a x t = 10 v x t + 1 - v x t, a y t = 10 v y t + 1 - v y t$	2
SVF	$Δ x t i, Δ y t i$	周围车辆i相对于目标车辆的横向与纵向位移，由公式可得： $Δ x t i = x t i - x t, Δ y t i = y t i - y t$	12
	$v x t i, v y t i$	周围车辆i的横向与纵向绝对速度，通过其位移可得： $v x t i = 10 x t + 1 i - x t i, v y t i = 10 y t + 1 i - y t i$	12
	$a x t, a y t$	周围车辆i的横向与纵向加速度，通过其速度可得： $a x t i = 10 v x t + 1 i - v x t i, a y t i = 10 v y t + 1 i - v y t i$	12
RF	$R l, R r$	$R l 、 R r$ 表示目标车辆左右车道标识位，如果其存在置为1，否则为0	2

表2

图2

图3

图4

图5

表3

表4

表5

图6

表6

表7

参考文献 21

1	Xing Y, Lyu C, Wang H, et al. Driver lane change intention inference for intelligent vehicles: framework, survey, and challenges[J]. IEEE Transactions on Vehicular Technology, 2019, 68(5): 4377-4390.
2	Houenou A, Bonnifait P, Cherfaoui V, et al. Vehicle trajectory prediction based on motion model and maneuver recognition[C]∥ IEEE/RSJ International Conference on Intelligent Robots and Systems, Tokyo, Japan, 2013: 4363-4369.
3	谢枫, 娄静涛, 赵凯, 等. 基于行为识别和曲率约束的车辆轨迹预测方法研究[J]. 汽车工程, 2019, 41(9): 1036-1042.
	Xie Feng, Lou Jing-tao, Zhao Kai, et al. A research on vehicle trajectory prediction method based on behavior recognition and curvature constraints[J]. Automotive Engineering, 2019, 41(9): 1036-1042.
4	Tsogas M, Polychronopoulos A, Floudas N, et al. Situation refinement for vehicle maneuver identification and driver's intention prediction[C]∥ 10th International Conference on Information Fusion, Quebec City, Canada, 2007: No.4408203.
5	Wang X, Guo Y, Bai C, et al. Driver's intention identification with the involvement of emotional factors in two-lane roads[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(11): 6866-6874.
6	祝俪菱, 刘澜, 赵新朋, 等. 基于支持向量机的车辆驾驶行为识别研究[J]. 交通运输系统工程与信息, 2017, 17(1): 91-97.
	Zhu Li-ling, Liu Lan, Zhao Xin-peng, et al. Driver behavior recognition based on support vector machine[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(1): 91-97.
7	Berndt H, Emmert J, Dietmayer K. Continuous driver intention recognition with hidden markov models[C]∥ 11th International IEEE Conference on Intelligent Transportation Systems, Beijing, China, 2008: 1189-1194.
8	Liu Q, Xu S, Lu C, et al. Early recognition of driving intention for lane change based on recurrent hidden semi-Markov model[J]. IEEE Transactions on Vehicular Technology, 2020, 69(10): 10545-10557.
9	刘志强, 吴雪刚, 倪捷, 等. 基于HMM和SVM级联算法的驾驶意图识别[J]. 汽车工程, 2018, 40(7): 858-864.
	Liu Zhi-qiang, Wu Xue-gang, Ni Jie, et al. Driving intention recognition based on hmm and svm cascade algorithm[J]. Automotive Engineering, 2018, 40(7): 858-864.
10	Fang H, Liu L, Gu Q, et al. Driving intention recognition of human drivers in mixed traffic flow[C]∥ IEEE 25th International Conference on Intelligent Transportation Systems, Macao, China, 2022: 153-157.
11	季学武, 费聪, 何祥坤, 等. 基于LSTM网络的驾驶意图识别及车辆轨迹预测[J]. 中国公路学报, 2019, 32(6): 34-42.
	Ji Xue-wu, Fei Cong, He Xiang-kun, et al. Intention recognition and trajectory prediction for vehicles using lstm network[J]. China Journal of Highway and Transport, 2019, 32(6): 34-42.
12	Chandra R, Guan T, Panuganti S, et al. Forecasting trajectory and behavior of road-agents using spectral clustering in graph-lstms[J]. IEEE Robotics and Automation Letters, 2020, 5(3): 4882-4890.
13	赵健, 宋东鉴, 朱冰, 等. 数据机理混合驱动的交通车意图识别方法[J]. 汽车工程, 2022, 44(7): 997-1008.
	Zhao Jian, Song Dong-jian, Zhu Bing, et al. Traffic vehicles intention recognition method driven by data and mechanism hybrid[J]. Automotive Engineering, 2022, 44(7): 997-1008.
14	赵建东, 赵志敏, 屈云超, 等. 轨迹数据驱动的车辆换道意图识别研究[J]. 交通运输系统工程与信息, 2022, 22(4): 63-71.
	Zhao Jian-dong, Zhao Zhi-min, Qu Yun-chao, et al. Vehicle lane change intention recognition driven by trajectory data[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(4): 63-71.
15	Guo H, Keyvan E M, Xie K. Lane change detection and prediction using real-world connected vehicle data[J]. Transportation Research Part C: Emerging Technologies, 2022, 142: No.103785.
16	Zhang Y, Shi X, Zhang S, et al. A xgboost-based lane change prediction on time series data using feature engineering for autopilot vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(10): 19187-19200.
17	Chen T, Guestrin C. Xgboost: a scalable tree boosting system[C]∥ 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, USA, 2016: 785-794.
18	Hu Y, Zhan W, Tomizuka M. Probabilistic prediction of vehicle semantic intention and motion[C]∥ IEEE Intelligent Vehicles Symposium (IV), Suzhou, China, 2018: 307-313.
19	Ding W, Chen J, Shen S. Predicting vehicle behaviors over an extended horizon using behavior interaction network[C]∥ International Conference on Robotics and Automation, Montreal, Canada, 2019: 8634-8640.
20	Altché F, Fortelle A. An LSTM network for highway trajectory prediction[C]∥ IEEE 20th International Conference on Intelligent Transportation Systems, Yokohama, Japan, 2017: 353-359.
21	田彦涛, 黄兴, 卢辉遒, 等. 基于注意力与深度交互的周围车辆多模态行为轨迹预测[J]. 吉林大学学报: 工学版, 2023, 53(5): 1474-1480.
	Tian Yan-tao, Huang Xing, Lu Hui-qiu, et al. Vehicle multi-modal maneuvers and trajectory prediction based on attention mechanism with deep interaction[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(5): 1474-1480.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

编号	特征名	特征描述	单位
1	Vehicle_ID	车辆编号	-
2	Frame_ID	帧数编号	-
3	Total_Frames	车辆在数据集中总帧数	-
4	Global_Time	全局时间（毫秒数）	ms
5	Local_X	横向相对位移	feet
6	Local_Y	纵向相对位移	feet
7	Lane_ID	车道编号	-

模型	未来轨迹/s
模型	1	2	3
MHA-LSTM	0.41	1.01	1.74
MTF-LSTM	0.48	0.89	1.47

历史轨迹长度/s	预测轨迹长度/s
历史轨迹长度/s	1	2	3
1	0.961	0.966	0.969
2	0.965	0.970	0.972
3	0.968	0.972	0.975
4	0.971	0.974	0.977
5	0.971	0.973	0.977

指标	模型	驾驶意图
指标	模型	左换道	车道保持	右换道
精确率	XGBoost	0.973	0.960	0.975
	S-XGBoost	0.950	0.963	0.948
	LSTM-XGBoost	0.984	0.965	0.983
召回率	XGBoost	0.971	0.959	0.978
	S-XGBoost	0.953	0.943	0.963
	LSTM-XGBoost	0.976	0.976	0.980
F₁ Score	XGBoost	0.972	0.959	0.977
	S-XGBoost	0.952	0.953	0.955
	LSTM-XGBoost	0.980	0.971	0.981
准确率	XGBoost	0.969
	S-XGBoost	0.953
	LSTM-XGBoost	0.977

[1]	陈永恒,杨家伟,孙经宇. 借道左转交叉口的网联左转车辆最佳轨迹控制[J]. 吉林大学学报(工学版), 2025, 55(2): 614-622.
[2]	金盛,李博林,薛炜. 智能网联车借用公交专用道的轨迹与信号协同优化[J]. 吉林大学学报(工学版), 2025, 55(2): 566-576.
[3]	李怀鑫,晏长根,林斌,石玉玲. 黏土-混凝土组合体强度及全过程统计损伤模型[J]. 吉林大学学报(工学版), 2025, 55(1): 245-255.
[4]	吴飞,王鹏程,杨康. 基于PCA-SSA-XGBoost的车辆驾驶性评估[J]. 吉林大学学报(工学版), 2025, 55(1): 105-115.
[5]	姚荣涵,祁文彦,胡宏宇,杜筱婧,乔延峰,王立冰. 考虑公交-合乘车道的多车道元胞自动机模型[J]. 吉林大学学报(工学版), 2025, 55(1): 162-174.
[6]	周荣贵,高沛,李雨璇,周建. 基于轨迹数据的高速公路小客车异常驾驶行为[J]. 吉林大学学报(工学版), 2024, 54(9): 2581-2587.
[7]	余萍,赵康,曹洁. 基于优化A-BiLSTM的滚动轴承故障诊断[J]. 吉林大学学报(工学版), 2024, 54(8): 2156-2166.
[8]	周求湛,冀泽宇,王聪,荣静. 基于在线压缩重构的非侵入式电力负荷监测[J]. 吉林大学学报(工学版), 2024, 54(6): 1796-1806.
[9]	马庆禄,闫浩,聂振宇,李杨梅. 匝道合流区智能网联车辆协同控制方法[J]. 吉林大学学报(工学版), 2024, 54(5): 1332-1346.
[10]	许清津,付锐,郭应时,吴付威. 载货汽车弯道侧翻路侧预测方法[J]. 吉林大学学报(工学版), 2024, 54(5): 1302-1310.
[11]	蒲云,徐银,刘海旭,谭一帆. 考虑多车影响的智能网联车跟驰模型[J]. 吉林大学学报(工学版), 2024, 54(5): 1285-1292.
[12]	曹倩,李志慧,陶鹏飞,李海涛,马永建. 面向交通事故检测及预防的异质传感器布设方法[J]. 吉林大学学报(工学版), 2024, 54(4): 969-978.
[13]	张鑫,胡启洲,何君,吴啸宇. 考虑交通梗塞的合流区交通状况诊断[J]. 吉林大学学报(工学版), 2024, 54(2): 478-484.
[14]	张卫华,刘嘉茗,解立鹏,丁恒. 网联混合环境快速路交织区自动驾驶车辆换道模型[J]. 吉林大学学报(工学版), 2024, 54(2): 469-477.
[15]	温惠英,何梓琦,李秋灵,赵胜. 高速公路货车换道冲突预测及其影响因素分析[J]. 吉林大学学报(工学版), 2024, 54(10): 2827-2836.