基于Bi-GLSTM网络的车辆驾驶意图分析与识别

doi:10.3901/JME.2024.10.051

摘要/Abstract

摘要： 驾驶意图识别能有效提高自车对其他交通参与者的轨迹预测能力，是实现智能车自主决策和规划的基础。然而动态复杂交通环境下周围车辆的交互是实现准确可靠驾驶意图识别亟待解决的挑战之一。为提高在动态复杂交通场景下驾驶意图识别的准确率，提出基于双向图长短时记忆网络(Bidirectional graph long short term memory, Bi-GLSTM)网络的驾驶意图时序识别模型。首先基于局部加权回归散点平滑法对原始数据集中的位置、速度和加速度进行平滑处理，并联合纵横向运动参数为数据标注驾驶意图；然后建立图注意力神经网络，分析和提取周围车辆与目标车辆之间的交互特征，嵌入注意力机制，分析周围车辆对目标驾驶意图的重要性，增强模型对相关性较大的车辆运动状态关注程度；融合周围车辆交互特征和目标车辆历史运动特征，为提高模型在动态复杂交通环境下的鲁棒性和可靠性，基于双向长短时记忆网络提取特征之间的时序特征；最后在公开数据集HighD上训练并验证模型的有效性，结果表明相比于图神经网络、循环神经网络等模型，识别准确率分别提高了11.33%、55.31%；通过可视化注意力权重，说明所提出的模型也一定程度上解决了可解释性问题。

关键词: 驾驶意图辨识, 智能网联汽车, 图神经网络

Abstract: Driving intention recognition is promised to effectively improve the vehicle’s ability to predict the trajectory of other traffic participants. However, the interaction of surrounding vehicles in a dynamic and complex traffic environment is one of the most challenges to be solved. In order to improve the accuracy of driving intention recognition in dynamic and complex traffics, a time-series recognition model of driving intention based on Bi-GLSTM network is proposed. First, the position, velocity and acceleration in the original dataset are smoothed based on the Local weighted regression scatter smoothing method. Also, the driving datasets are labeled with driving intention based on the longitudinal and lateral motion parameters. Subsequently, a graph attention neural network is established to extract interaction features among surrounding vehicles, where attention mechanism is embedded, to enhance highly related vehicle motion states. To further improve the robustness of the model in dynamic and complex traffics, Bidirectional long short-term memory network is used to extract deep temporal features of interaction and historical motion. Moreover, our model is trained and verified on the public datasets HighD. Compared with GNN and RNN model, the recognition accuracy increased by 11.33%, 55.31%. By visualizing the attention weight, it shows that the proposed model also solves the problem of explainability to a certain extent.

Key words: driver intention inference, connected and automated vehicles, graph neural network

中图分类号:

TG156

李琳, 赵万忠, 王春燕. 基于Bi-GLSTM网络的车辆驾驶意图分析与识别[J]. 机械工程学报, 2024, 60(10): 51-63.

LI Lin, ZHAO Wanzhong, WANG Chunyan. Driving Intention Recognition Model Based on Bi-GLSTM Network[J]. Journal of Mechanical Engineering, 2024, 60(10): 51-63.

参考文献

[1] DO J，HAN K，CHOI S B，et al. Inference of vehicle lane change intention using multiple model estimator in automated highway driving[C]//Proceedings of the 22nd International Conference on Control，Automation and Systems (ICCAS). Jeju:IEEE，2022:366-372.
[2] 张雁雁. 面向行人防碰撞预警的驾驶员驾驶意图辨识方法研究[D]. 大连:大连理工大学，2015. ZHANG Yanyan. Research on driver's driving intention identification method for pedestrian collision avoidance warning[D]. Dalian:Dalian University of Technology，2015.
[3] 武林. 基于GRU-ATT的车辆驾驶意图识别与轨迹预测[D]. 北京:中国矿业大学，2022. WU Lin. Vehicle driving intention recognition and trajectory prediction based on GRU-ATT[D]. Beijing:China University of Mining & Technology，2022.
[4] WU Z Q，LIANG K C，LIU D C，et al. Driver lane change intention recognition based on attention enhanced residual-MBi-LSTM network[J]. IEEE Access，2022，10:58050-58061.
[5] WANG K，HOU J，ZENG X L. Lane-change intention prediction of surrounding vehicles using BiLSTM-CRF models with rule embedding[C]//2022 China Automation Congress (CAC). Xiamen:IEEE. 2022:2764-2769.
[6] 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.
[7] YANG J H，KIM D J，CHUNG C C，et al. Lane change intention inference of surrounding vehicle:Comparative study on Relevance Vector Machine (RVM) and Support Vector Machine (SVM)[C]//Proceedings of the 21st International Conference on Control，Automation and Systems (ICCAS). Jeju:IEEE，2021:1580-1585.
[8] 宋晓琳，郑亚奇，曹昊天. 基于HMM-SVM的驾驶员换道意图辨识研究[J]. 电子测量与仪器学报，2016，30(1):58-65. SONG Xiaolin，ZHENG Yaqi，CAO Haotian. Research on driver lane-changing intention identification based on HMM-SVM[J]. Journal of Electronic Measurement and Instrumentation，2016，30(1):58-65.
[9] FANG H Z，LIU L，GU Q，et al. Driving intention recognition of human drivers in mixed traffic flow[C]//Proceedings of the IEEE 25th International Conference on Intelligent Transportation Systems (ITSC). Macau:IEEE，2022:153-157.
[10] 雷昊. 基于历史行为序列识别的环境车辆驾驶意图及轨迹预测[D]. 长沙:湖南大学，2021 LEI Hao. Surrounding vehicle driving intention and trajectory prediction based on historical behavior sequence recognition[D]. Changsha:Hunan University，2021.
[11] VELLENGA K，STEINHAUER H J，KARLSSON A，et al. Driver intention recognition:State-of-the-art review[J]. IEEE Open Journal of Intelligent Transportation Systems，2022，3:602-616.
[12] 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.
[13] LIU Y，ZHAO P，QIN D，et al. Driving intention identification based on long short-term memory and a case study in shifting strategy optimization[J]. IEEE Access，2019，7:128593-128605.
[14] 张运开. 深度学习在时间序列分类问题中的研究及应用[D]. 上海:华东师范大学，2021. ZHANG Yunkai. Research and application of deep learning in time series classification problems[D]. Shanghai:East China Normal University，2021.
[15] TANG L，WANG H，ZHANG W，et al. Driver lane change intention recognition of intelligent vehicle based on long short-term memory network[J]. IEEE Access，2020，8:136898-136905.
[16] LIU C Z，ALIAMANI H，KAYAKLI M，et al. Behavior-intention analysis and human-aware computing:Case study and discussion[C]//Proceedings of the 12th IEEE Conference on Industrial Electronics and Applications (ICIEA). Siem Reap:IEEE，2017:516-521.
[17] 高凯，李勋豪，胡林，等. 基于多头注意力的CNN-LSTM的换道意图预测[J]. 机械工程学报，2022，58(22):369-378. GAO Kai，LI Xunhao，HU Lin, et al. Lane-changing intention prediction using CNN-LSTM based on multi-head attention[J]. Journal of Mechanical Engineering，2022，58(22):369-378.
[18] HUANG H，ZENG Z N，YAO D Y，et al. Spatial-temporal ConvLSTM for vehicle driving intention prediction[J]. Tsinghua Science and Technology，2022，27(3):599-609.
[19] HAO Z，HUANG X，WANG K，et al. Attention-based gru for driver intention recognition and vehicle trajectory prediction[C]//Proceedings of the 20204th CAA International Conference on Vehicular Control and Intelligence (CVCI). Hangzhou:IEEE，2020:86-91.
[20] ROBERT K，JULIAN B，LAURENT K，et al. The highD dataset:A drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems[C]//Proceedings of the 201821st International Conference on Intelligent Transportation Systems (ITSC). Hawaii:IEEE，2018:2118-2125.
[21] PRESS W H，TEUKOLSKY S A. Savitzky‐Golay smoothing filters[J]. Computers in Physics，1990，4(6):669-672.
[22] PINTUS E，SORBOLINI S，ALBERA A，et al. Use of locally weighted scatterplot smoothing (lowess) regression to study selection signatures in piedmontese and italian brown cattle breeds[J]. Animal Genetics，2014，45(1):1-11.
[23] 杨诚. 基于隐马尔可夫理论的驾驶人换道意图识别研究[D]. 吉林:吉林大学，2016. YANG Cheng. Research on driver's lane-changing intention recognition based on hidden Markov theory[D]. Jilin:Jilin University，2016.
[24] 马晶晶. 基于隐马尔可夫理论的驾驶意图辨识研究[D]. 长沙:长沙理工大学，2012. MA Jingjing. Research on driving intention identification based on hidden Markov theory[D]. Changsha:Changsha University of Science and Technology，2012.
[25] 姚荣涵，徐文韬，郭伟伟. 基于因子长短期记忆的驾驶人接管行为及意图识别[J]. 吉林大学学报(工学版)，2023，53(3):758-771. YAO Ronghan，XU Wentao，GUO Weiwei. Driver takeover behavior and intention recognition based on factor long-short-term memory[J]. Journal of Jilin University (Engineering Edition)，2023，53(3):758-771.
[26] ZHOU J，CUI G，HU S，et al. Graph neural networks:A review of methods and applications[J]. AI Open，2020，1:57-81.