无人驾驶汽车周边车辆行为识别算法研究*

doi:10.19562/j.chinasae.qcgc.2020.11.003

摘要/Abstract

摘要： 周边车辆行为识别对于提升无人驾驶汽车决策规划的合理性和控制安全性至关重要。传统的周边车辆行为识别方法识别精度普遍不高,且缺乏对交通主体相互之间邻域影响的考虑,算法鲁棒性较差。针对此,本文中提出了一种SLSTMAT(Social-LSTM-Attention)算法,创新性地引入目标车辆社交特征并通过卷积神经网络提取,建立了基于深度学习的车辆行为识别模型,应用注意力机制来捕捉行为时窗中的多时步信息,实现了周边车辆行为准确识别。采用HighD轨迹数据集和实车数据进行算法验证。结果表明,所提算法对周边车辆行为识别的准确率达94.01%,在目标车辆到达换道点的前1　s时刻行为识别精度达90%,具有较高的工程应用价值。

关键词: 无人驾驶汽车, 行为识别, 长短时记忆网络, 注意力机制, 社交特征

Abstract: The behavior recognition of surrounding vehicles is very important to improve the decision-making rationality and the control safety for driverless cars. Traditional methods of surrounding vehicle’s behavior recognition suffer from low accuracy and poor robustness, without considering the interaction between adjacent traffic objects. In order to solve these problems, SLSTMAT (Social-LSTM-Attention) algorithm is proposed to achieve high accuracy for surrounding vehicle’s behavior recognition. The social characteristics of target vehicle are innovatively introduced and extracted by convolutional neural network. Based on deep learning,the recognition model for vehicle behavior is established. Attention mechanism is applied to capture multiple time-step information in the behavior time window. The algorithm is verified by HighD trajectory data set and real vehicle data. The results show that the accuracy rate of SLSTMAT for surrounding vehicle’s behavior recognition reaches 94.01%, and the precision of behavior recognition reaches 90% at 1s before the target vehicle driving to the lane change point, which has high engineering application value

Key words: driverless cars, behavior recognition, long and short term memory network, attention mechanism, social characteristics

蔡英凤, 邰康盛, 王海, 李祎承, 陈龙. 无人驾驶汽车周边车辆行为识别算法研究^*[J]. 汽车工程, 2020, 42(11): 1464-1472.

Cai Yingfeng, Tai Kangsheng, Wang Hai, Li Yicheng, Chen Long. Research on Behavior Recognition Algorithm of Surrounding Vehicles for Driverless Car[J]. Automotive Engineering, 2020, 42(11): 1464-1472.

参考文献

[1] XIE G, GAO H, QIAN L, et al. Vehicle trajectory prediction by integrating physics and maneuver-based approaches using interactive multiple models[J]. IEEE Transactions on Industrial Electronics, 2017, 65(7): 5999-6008.
[2] DING W, CHEN J, SHEN S. Predicting vehicle behaviors over an extended horizon using behavior interaction network[C]. 2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 8634-8640.
[3] 熊帅. 快速路汇入行为分析及合流区组织优化[D]. 长春:吉林大学, 2019.
XIONG S. Analysis of expressway merging behavior and optimization of merging area organization [D]. Changchun: Jilin University, 2019.
[4] 党彤. 基于贝叶斯网络的车辆变道行为分析[D]. 西安:西安理工大学, 2018.
DANG T. Lane changing behavior analysis of vehicles based on Bayesian network [D]. Xi'an: Xi'an University of Technology, 2018.
[5] 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, 2016, 17(1): 91- 97.
ZHU L, LIU L, ZHAO X, et al. Research on vehicle driving behavior recognition based on support vector machine [J]. Transportation System Engineering and Information, 2016, 17 (1): 91- 97.
[6] 李建平. 面向智能驾驶的交通车辆运动预测方法研究[D]. 长春:吉林大学, 2018.
LI J. Research on intelligent driving oriented traffic vehicle motion prediction method [D]. Changchun: Jilin University, 2018.
[7] GENG X, LIANG H, YU B, et al. A scenario-adaptive driving behavior prediction approach to urban autonomous driving[J]. Applied Sciences, 2017, 7(4): 426-447.
[8] XU D, HE X, ZHAO H, et al. Ego-centric traffic behavior understanding through multi-level vehicle trajectory analysis[C]. 2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017: 211-218.
[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(8):858-864.
LIU Z, WU X, NI J, et al. Driving intention recognition based on HMM and SVM cascade algorithm [J]. Automotive Engineering, 2018,40(8): 858-864.
[10] 李创. 智能汽车对前方车辆的运动感知与换道意图辨识[D]. 西安:西安理工大学, 2018.
LI C. Motion perception and lane change intention identification for intelligent vehicles [D]. Xi'an: Xi'an University of Technology, 2018.
[11] DENG Q, WANG J, SOFFKER D. Prediction of human driver behaviors based on an improved HMM approach[C]. 2018 IEEE Intelligent Vehicles Symposium (IV). IEEE, 2018: 2066-2071.
[12] LI K, WANG X, XU Y, et al. Lane changing intention recogni-tion based on speech recognition models[J]. Transportation Research Part C: Emerging Technologies, 2016, 69: 497-514.
[13] 王一鸣, 陈恳. 基于稀疏DBN和双向LSTM的视觉语音识别算法[J]. 数据通信, 2019(3) :9-15.
WANG Y, CHEN K. Visual speech recognition algorithm based on sparse DBN and bidirectional LSTM [J]. Data Communica-tion, 2019(3): 9-15.
[14] HUANG X, TAN H, LIN G, et al. A LSTM-based bidirectional translation model for optimizing rare words and terminologies[C]. 2018 International Conference on Artificial Intelligence and Big Data (ICAIBD). IEEE, 2018: 185-189.
[15] 姚朝, 辛平安, 施卜今, 等. 基于 LSTM 时间递归神经网络的短期电力负荷预测[J]. 云南水力发电, 2019,35(3): 163- 165.
YAO C, XIN P, SHI B, et al. Short term power load forecasting based on LSTM time recurrent neural network [J]. Yunnan Hydropower, 2019,35(3): 163-165.
[16] 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, FEI C, HE X, et al. Driving intention recognition and vehicle trajectory prediction based on LSTM network [J]. China Journal of Highway and Transport, 2019, 32 (6): 34-42.
[17] ALAHI A, GOEL K, RAMANATHAN V, et al. Social LSTM: human trajectory prediction in crowded spaces[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 961-971.
[18] KRAJEWSKI R, BOCK J, KLOEKER L, et al. The HighD dataset: a drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C]. 2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2018: 2118-2125.

无人驾驶汽车周边车辆行为识别算法研究^*

Research on Behavior Recognition Algorithm of Surrounding Vehicles for Driverless Car

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

Metrics

本文评价

推荐阅读 10

[1]	胡宏宇, 黎烨宸, 张争光, 曲优, 何磊, 高镇海. 基于多尺度骨架图和局部视觉上下文融合的驾驶员行为识别方法[J]. 汽车工程, 2024, 46(1): 1-8.
[2]	李勇滔,孙晨旭,郑伟光,许恩永,李育方,王善超. 基于毫米波雷达与视觉融合的碰撞预警[J]. 汽车工程, 2023, 45(9): 1666-1676.
[3]	赵霞,李朝,付锐,葛振振,王畅. 基于深度卷积-Tokens降维优化视觉Transformer的分心驾驶行为实时检测[J]. 汽车工程, 2023, 45(6): 974-988.
[4]	李琳辉,张鑫亮,付一帆,连静,马家旭. 基于TC-YOLOv7算法的可见光与红外后融合检测研究[J]. 汽车工程, 2023, 45(12): 2280-2290.
[5]	陈龙,杨晨,蔡英凤,王海,李祎承. 基于多模态特征融合的行人穿越意图预测方法[J]. 汽车工程, 2023, 45(10): 1779-1790.
[6]	张小俊,奚敬哲,史延雷,袁安录. 面向路侧视角目标检测的轻量级YOLOv7-R算法[J]. 汽车工程, 2023, 45(10): 1833-1844.
[7]	张斌,付俊怡,夏金祥. 基于类间距优化的分心驾驶行为识别模型训练方法[J]. 汽车工程, 2022, 44(2): 225-232.
[8]	王大方,尚海,曹江,王涛,夏祥腾,韩雨霖. 基于自注意力机制的自动驾驶场景点云语义分割方法[J]. 汽车工程, 2022, 44(11): 1656-1664.
[9]	蔡英凤,吕志军,孙晓强,王海,刘擎超,陈龙,袁朝春. 基于并线行为识别的自适应巡航控制方法[J]. 汽车工程, 2021, 43(7): 1077-1087.
[10]	刘军,陈岚磊,李汉冰. 基于类人视觉的多任务交通目标实时检测模型[J]. 汽车工程, 2021, 43(1): 50-58.
[11]	白中浩, 王韫宇, 张林伟. 基于图卷积网络的多信息融合驾驶员分心行为检测^*[J]. 汽车工程, 2020, 42(8): 1027-1033.
[12]	彭晓燕, 谢浩, 黄晶. 无人驾驶汽车局部路径规划算法研究^*[J]. 汽车工程, 2020, 42(1): 1-10.
[13]	边辰通, 殷国栋, 徐利伟, 张宁, 朱侗. 无人驾驶汽车编队雾天主动拯救系统^*[J]. 汽车工程, 2020, 42(1): 20-26.
[14]	谢枫, 娄静涛, 赵凯, 齐尧. 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.