Collision avoidance model and its validation for intelligent vehicles based on deep learning LSTM

doi:10.3969/j.issn.1674-8484.2022.01.010

Abstract

Abstract:

A collision avoidance model was established for intelligent vehicles based on a deep learning Long Short-Term Memory (LSTM) network. Some driving simulation experiments were carried out at the conditions of following car dangerous driving or front car emergency braking, through a driving simulation platform with a simulation software of Virtual Test Driving (VTD). The relative distances, the relative speeds, the decelerations of the preceding car, the collision times, and the lateral distances were taken as the input parameters. Model mobility was verified by an untrained sample data, and compared with a traditional Back Propagation (BP) neural network. The results show that the R² values of the model are higher than that of the traditional BP model by 0.17 and 0.21, respectively for the predicational decelerations and the predicational steering wheel angles. Therefore, this model has a fitting goodness for the big data samples and can represent the driver actual collision-avoidance behavior, with a value in promoting application of driver assistance system.

Key words: intelligent vehicles, collision avoidance, driving simulators, deep learning, long short-term memory (LSTM), driver assistance systems

CLC Number:

TP183
U491.2

FANG Liang, GUAN Zhiwei, WANG Tao, GONG Jinfeng, DU Feng. Collision avoidance model and its validation for intelligent vehicles based on deep learning LSTM[J]. Journal of Automotive Safety and Energy, 2022, 13(1): 104-111.

Figures/Tables 16

References 18

[1]	孙维富. 基于数据挖掘的高速公路交通事故分析及预防对策研究[D]. 吉林: 吉林大学, 2018: 14-15.
	SUN Weifu. Research on highway traffic accident analysis and preventive countermeasures based on data mining[D]. Jilin: Jilin University, 2018: 14-15. (in Chinese)
[2]	何仁, 冯海鹏. 自动紧急制动(AEB)技术的研究与进展[J]. 汽车安全与节能学报, 2019,10(1):1-15.
	HE Ren, FENG Haipeng. Research and development of autonomous emergency brake (AEB) technology[J]. J Automotive Safe Energy, 2019,10(1):1-15. (in Chinese)
[3]	刘志强, 张春雷, 张爱红, 等. 基于驾驶行为的追尾避撞控制策略研究[J]. 汽车工程, 2017, 39(9): 1068-1073+1080.
	LIU Zhiqiang, ZHANG Chunlei, ZHANG Aihong, et al. A Study on the control strategy for Rear-end collision avoidance based on drivers’ behavior[J]. Automotive Engineering, 2017, 39(9): 1068-1073+1080. (in Chinese)
[4]	PENG Tao, SU Lili, ZHANG Ronghui, et al. A new safe lane-change trajectory model and collision avoidance control method for automatic driving vehicles[J]. Expert Syst Appl, 2020, 141: 112953.1-112953.17.
[5]	Dogan U, Edelbrunner J, Iossifidis I. Autonomous driving: a comparison of machine learning techniques by means of the prediction of lane change behavior[C]// IEEE Int’l Conf Robotics Biomimetics, New York, 2011: 1837-1843.
[6]	王俊彦, 蔡骏宇. 基于RBF神经网络的车辆安全换道时机决策模型研究[J]. 重庆理工大学学报(自然科学版), 2019,33(11):47-51, 80.
	WANG Junyan, CAI Junyu. Research on modelling vehicle safety lane-changing timing decision based on RBF neural network[J]. J Chongqing Univ Techn (Natu Sci), 2019,33(11):47-51, 80. (in Chinese)
[7]	Motamedidehkordi N, Amini S, Hoffmann S, et al. Modeling tactical lane-change behavior for automated vehicles: A supervised machine learning approach[C]// IEEE Int’l Conf’Models Tech Intell Transp Syst, 2017: 268-273.
[8]	刘志强, 王玲, 贾海江, 等. 基于BP神经网络的驾驶员制动行为模型研究[J]. 机械设计与制造, 2019(6):37-41.
	LIU Zhiqiang, WANG Ling, JIA Haijiang, et al. Driver’s braking behavior based on BP neural network[J]. Mach Desi Manuf, 2019(6):37-41. (in Chinese)
[9]	黄玲, 郭亨聪, 张荣辉, 等. 人机混驾环境下基于LSTM的无人驾驶车辆换道行为模型[J]. 中国公路学报, 2020,33(7):156-166.
	HUANG Ling, GUO Hengcong, ZHANG Ronghui, et al. LSTM-based lane-changing behavior model for unmanned vehicle under environment of heterogeneous Human-driven and autonomous vehicles[J]. China J Highw Transp, 2020,33(7):156-166. (in Chinese)
[10]	XIE Dongfan, FANG Zhezhe, JIA Bin, et al. A data-driven lane-changing model based on deep learning[J]. Transp Res Part C Emerging Tech, 2019,106:41-60.
[11]	张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021,57(11):57-69.
	ZHANG Chi, GUO Yuan, LI Ming. Review of development and application of artificial neural network models[J]. Computer Eng Appl, 2021,57(11):57-69. (in Chinese)
[12]	马焱棋, 林群, 赵昱程, 等. 基于深度学习LSTM对交通流状态的预测[J]. 数学的实践与认识, 2021,51(04):47-56.
	MA Yanqi, LIN Qun, ZHAO Yucheng, et al. Prediction of traffic flow state based on deep learning LSTM[J]. Math Practice Theory. 2021,51(4):47-56. (in Chinese)
[13]	杨丽, 吴雨茜, 王俊丽, 等. 循环神经网络研究综述[J]. 计算机应用, 2018, 38(S2):1-6+26.
	YANG Li, WU Yuqian, WANG Junli, et al. Research on recurrent neural network [J]. J Computer Appl, 2018, 38(S2): 1-6+26. (in Chinese)
[14]	郭应时, 付锐, 赵凯, 等. 驾驶人换道意图实时识别模型评价及测试[J]. 吉林大学学报(工学版), 2016,46(6):1836-1844.
	GUO Yingshi, FU Rui, ZHAO Kai, et al. Evaluation and test of real-time identification models of driver’s lane change intention[J]. J Jilin Univ (Eng Tech Ed), 2016,46(6):1836-1844. (in Chinese)
[15]	HUANG Ling, GUO Hengcong, ZHANG Ronghui, et al. Capturing drivers’ lane changing behaviors on operational level by data driven methods[J]. IEEE Access, 2018,6:57497-57506.
[16]	ZHENG Zhao, CHEN Weihai, WU Xingming, et al. LSTM network: A deep learning approach for short-term traffic forecast[J]. Iet’l Intell Transp Syst, 2017,11(2):68-75.
[17]	王克杰, 张瑞. 基于改进BP神经网络的短期电力负荷预测方法研究[J]. 电测与仪表, 2019,56(24):115-121.
	WANG Kejie, ZHANG Rui. Research on short-term power load forecasting method based on improved BP neural network[J]. Elect Mea Instr, 2019,56(24):115-121. (in Chinese)
[18]	康梦轩, 宋俊平, 范鹏飞, 等. 基于深度学习的网络流量预测研究综述[J]. 计算机工程与应用, 2021,57(10):1-9.
	KANG Mengxuan, SONG Junping, FAN Pengfei, et al. Survey of network traffic forecast based on deep learning[J]. Computer Engi Appl, 2021,57(10):1-9. (in Chinese)

编号	性别	年龄/岁	驾龄/年	编号	性别	年龄/岁	驾龄/年
1	男	24	4	11	男	24	3
2	男	24	3	12	男	24	4
3	男	26	5	13	女	26	5
4	男	24	4	14	女	24	4
5	男	30	6	15	女	23	5
6	男	23	3	16	女	24	3
7	男	28	5	17	女	35	10
8	男	25	4	18	女	34	9
9	男	37	10	19	女	24	3
10	男	38	17	20	女	26	5

编号	性别	年龄/岁	驾龄/年	编号	性别	年龄/岁	驾龄/年
1	男	24	4	11	男	24	3
2	男	24	3	12	男	24	4
3	男	26	5	13	女	26	5
4	男	24	4	14	女	24	4
5	男	30	6	15	女	23	5
6	男	23	3	16	女	24	3
7	男	28	5	17	女	35	10
8	男	25	4	18	女	34	9
9	男	37	10	19	女	24	3
10	男	38	17	20	女	26	5

t	δ_sw	γ_br	γ_ga	v_xexp	a_xexp	v_xob	a_xob	X_exp	Y_exp	X_ob	Y_ob
ms	(°)	%	%	m·s^-1	m·s^-2	m·s^-1	m·s^-2	m	m	m	m
0	0.001	0.008	0.000	23.305	-0.259	18.222	-5.000	6 316.583	0.076	6 394.985	0.000
17	0.001	0.051	0.000	23.300	-0.333	18.139	-5.000	6 316.971	0.075	6 395.288	0.000
33	0.006	0.083	0.000	23.291	-0.745	18.056	-5.000	6 317.359	0.074	6 395.590	0.000
50	0.006	0.106	0.000	23.276	-1.044	17.972	-5.000	6 317.747	0.074	6 395.890	0.000
67	0.006	0.138	0.000	23.257	-1.268	17.889	-5.000	6 318.135	0.073	6 396.189	0.000
83	0.009	0.177	0.000	23.233	-1.567	17.806	-5.000	6 318.523	0.072	6 396.486	0.000
100	0.013	0.008	0.000	23.305	-0.259	18.222	-5.000	6 316.583	0.076	6 394.985	0.000

t	δ_sw	γ_br	γ_ga	v_xexp	a_xexp	v_xob	a_xob	X_exp	Y_exp	X_ob	Y_ob
ms	(°)	%	%	m·s^-1	m·s^-2	m·s^-1	m·s^-2	m	m	m	m
0	0.001	0.008	0.000	23.305	-0.259	18.222	-5.000	6 316.583	0.076	6 394.985	0.000
17	0.001	0.051	0.000	23.300	-0.333	18.139	-5.000	6 316.971	0.075	6 395.288	0.000
33	0.006	0.083	0.000	23.291	-0.745	18.056	-5.000	6 317.359	0.074	6 395.590	0.000
50	0.006	0.106	0.000	23.276	-1.044	17.972	-5.000	6 317.747	0.074	6 395.890	0.000
67	0.006	0.138	0.000	23.257	-1.268	17.889	-5.000	6 318.135	0.073	6 396.189	0.000
83	0.009	0.177	0.000	23.233	-1.567	17.806	-5.000	6 318.523	0.072	6 396.486	0.000
100	0.013	0.008	0.000	23.305	-0.259	18.222	-5.000	6 316.583	0.076	6 394.985	0.000

自车减速度		方向盘转角
误差区间/ (m·s^-2)	误差分布/ %	误差区间/ mrad	误差分布/ %
[-0.2,0.2]	48.69	[-60, 60]	64.60
[-0.6,0.6]	78.27	[-120, 120]	85.55
[-1.2,1.2]	93.31	[-240, 240]	96.99