Lane Change Intention Recognition Based on Multi-Model Fusion

doi:10.3778/j.issn.1002-8331.2210-0058

Abstract

Abstract: Fast and accurate identification of lane-changing intentions of surrounding vehicles is of great significance for decision support and safety prevention of advanced autonomous driving assistance systems. Aiming at the problems that the existing methods fail to fully consider the interaction between vehicles and the front-to-back dependence of trajectory data, this paper proposes a lane-changing intention recognition framework based on multi-model fusion. The framework mainly includes input processing and lane-changing intention recognition. The input processing part cleans, labels, slices and one-hot codes the vehicle track data. The BiLSTM-F（BiLSTM-fusion） model is specifically proposed for lane change intention recognition. In this model, the attention mechanism is introduced into BiLSTM, and the weight of the output information in the input processing part is divided. Finally, conditional random field is introduced to fully learn the dependency of input data and quickly find the global optimal lane-changing intention. In the experiment, NGSIM is used for training and evaluation. The validation results show that the model can achieve the highest accuracy of 97.19%, and can identify the vehicle’s lane changing intention 2 s before the vehicle arrives at the lane-changing point, with an accuracy of 94.16%. Compared with the baseline lane-change intention recognition model, the accuracy, loss, F1 value and stability of the proposed model are better than those of the baseline model.

Key words: intelligent transportation, lane change intention recognition, anterior and posterior dependence, attention mechanism, conditional random field

摘要： 快速准确识别出周围车辆的换道意图对高级自动驾驶辅助系统的决策支持和安全预防具有重要意义。针对现有方法未能充分考虑车辆之间的交互作用以及轨迹数据的前后依赖性问题，提出一种基于多模型融合的换道意图识别框架。该换道意图识别框架主要包括输入处理与换道意图识别两部分。输入处理部分对车辆轨迹数据进行清洗、贴标、切片以及one-hot编码。换道意图识别部分则具体提出BiLSTM-F（BiLSTM-fusion）模型，在该模型中将注意力机制（attention mechanism）引入双向长短期记忆网络（BiLSTM），对输入处理部分的输出信息进行权重划分，使模型能将注意力更加集中于对换道意图影响较大的特征信息上，最后引入条件随机场（conditional random field），充分学习输入数据的前后依赖性并快速找出全局最优的换道意图。实验中使用公开数据集NGSIM进行训练并评估，验证结果表明该模型的准确率最高能达到97.19%，并且可在车辆到达换道点前2?s识别车辆的换道意图，准确率为94.16%。与基线换道意图识别模型相比，所提出模型的准确率、损失、F1值和稳定性均优于基线模型。

关键词: 智能交通, 换道意图识别, 前后依赖性, 注意力机制, 条件随机场

FANG Yijie, LIAO Zhuhua, HUANG Haokai, LI Yanjun. Lane Change Intention Recognition Based on Multi-Model Fusion[J]. Computer Engineering and Applications, 2024, 60(2): 344-352.

方艺洁, 廖祝华, 黄浩楷, 李彦君. 多模型融合的换道意图识别研究[J]. 计算机工程与应用, 2024, 60(2): 344-352.

References

[1] 郑赟. 2020中国智能电动汽车发展报告[R]. 广州: 罗兰贝格, 2020.
ZHENG Y. 2020 China intelligent electric vehicle development report[R]. Guangzhou: Roland Berger, 2020.
[2] LITMAN T. Autonomous vehicle implementation predictions: implications for transport planning[C]//Transportation Research Board Annual Meeting, 2015.
[3] LEFèVRE S, VASQUEZ D, LAUGIER C. A survey on motion prediction and risk assessment for intelligent vehicles[J]. ROBOMECH Journal, 2014, 1(1): 1-14.
[4] LI G, LI S E, LIAO Y, et al. Lane change maneuver recognition via vehicle state and driver operation signals—results from naturalistic driving data[C]//2015 IEEE Intelligent Vehicles Symposium (IV), 2015: 865-870.
[5] KUMAR P, PERROLLAZ M, LEFEVRE S, et al. Learning-based approach for online lane change intention prediction[C]//2013 IEEE Intelligent Vehicles Symposium (IV), 2013: 797-802.
[6] MORRIS B, DOSHI A, TRIVEDI M. Lane change intent prediction for driver assistance: on-road design and evaluation[C]//2011 IEEE Intelligent Vehicles Symposium (IV), 2011: 895-901.
[7] DING J, DANG R, WANG J, et al. Driver intention recognition method based on comprehensive lane-change environment assessment[C]//2014 IEEE Intelligent Vehicles Symposium Proceedings, 2014: 214-220.
[8] ZONG C, WANG C, YANG D, et al. Driving intention identification and maneuvering behavior prediction of drivers on cornering[C]//2009 International Conference on Mechatronics and Automation, 2009: 4055-4060.
[9] CHEN C, LIU L, QIU T, et al. Driver’s intention identification and risk evaluation at intersections in the Internet of vehicles[J]. IEEE Internet of Things Journal, 2018, 5(3): 1575-1587.
[10] WANG S, Fujii H, Yoshimura S. Accurate and efficient driving intention inference based on traffic environment information and FES-XGB framework[J]. Journal of Information Processing, 2022, 30: 30-41.
[11] WU Z, LIANG K, LIU D, et al. Driver lane change intention recognition based on attention enhanced residual-MBi-LSTM network[J]. IEEE Access, 2022, 10: 58050-58061.
[12] LI L, ZHAO W, XU C, et al. Lane-change intention inference based on RNN for autonomous driving on highways[J]. IEEE Transactions on Vehicular Technology, 2021, 70(6): 5499-5510.
[13] XU G, LI L, SONG Z. Driver behavior analysis based on Bayesian network and multiple classifiers[C]//IEEE International Conference on Intelligent Computing & Intelligent Systems, 2010.
[14] 毕胜强, 梅德纯, 刘志强, 等. 面向驾驶行为预警的换道意图辨识模型研究[J]. 中国安全科学学报, 2016, 26(2): 91-95.
BI S Q, MEI D C, LIU Z Q, et al. Research on lane change intention identification model for driving behavior warning [J]. China Safety Science Journal, 2016, 26(2): 91-95.
[15] MCCALL J C, TRIVEDI M M, WIPF D, et al. Lane change intent analysis using robust operators and sparse bayesian learning[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), 2006.
[16] 姜顺明, 匡志豪, 王奕轩, 等. 基于贝叶斯网络的车辆并线意图识别[J]. 河南科技大学学报 (自然科学版), 2021, 42(5): 39-44.
JIANG S M, KUANG Z H, WANG Y X, et al. Vehicles and lines based on Bayesian network intention recognition[J]. Journal of Henan University of Science and Technology (Natural Science Edition), 2021, 42 (5): 39-44.
[17] 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 Z Q, WU X G, NI J, et al. Driving intention recognition based on HMM and SVM cascade algorithm[J]. Automotive Engineering, 2018, 40 (7): 858-864.
[18] 张海伦, 付锐. 高速场景相邻前车驾驶行为识别及意图预测[J]. 交通运输系统工程与信息, 2020, 20(1): 40-46.
ZHANG H L, FU R. Driving behavior recognition and intention prediction of adjacent front vehicles in high-speed scene[J]. Transportation System Engineering and Information, 2020, 20(1): 40-46.
[19] 王亚伦, 陈焕明, 赵岩. 基于条件随机场的驾驶意图识别研究[J]. 青岛大学学报 (工程技术版), 2021, 36(4): 88-94.
WANG Y L, CHEN H M, ZHAO Y. Research on driving intention recognition based on conditional random field[J]. Journal of Qingdao University (Engineering and Technology Edition), 2021, 36(4): 88-94.
[20] 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 X W, FEI C, HE X K, et al. LSTM network based driver intention recognition and vehicle track prediction[J]. China Journal of Highway and Transport, 2019, 32(6): 34-42.
[21] 宋威龙. 城区动态环境下智能车辆行为决策研究[D]. 北京: 北京理工大学, 2016.
SONG W L. Research on intelligent vehicle behavior decision in urban dynamic environment[D]. Beijing: Beijing Institute of Technology, 2016.
[22] 宋晓琳, 曾艳兵, 曹昊天, 等. 基于长短期记忆网络的换道意图识别方法[J]. 中国公路学报, 2021, 34(11): 236-245.
SONG X L, ZENG Y B, CAO H T, et al. Lane change intention recognition method based on long short-term memory network[J]. China Journal of Highway and Transport, 2021, 34(11): 236-245.
[23] LAFFERTY J, MCCALLUM A, PEREIRA F. Conditional random fields: probabilistic models for segmenting and labeling sequence data[C]//Proceedings of the 18th International Conference on Machine Learning, 2001.
[24] XIA Y, QU Z, SUN Z, et al. A human-like model to understand surrounding vehicles’ lane changing intentions for autonomous driving[J]. IEEE Transactions on Vehicular Technology, 2021, 70(5): 4178-4189.
[25] JIN H, DUAN C, LIU Y, et al. Gauss mixture hidden Markov model to characterise and model discretionary lane‐change behaviours for autonomous vehicles[J]. IET Intelligent Transport Systems, 2020, 14(5): 401-411.