基于驾驶风格的前撞预警系统报警策略

doi:10.19562/j.chinasae.qcgc.2021.03.014

摘要/Abstract

摘要：

考虑驾驶风格的影响，设计了一种优化的纵向相对距离预测模型，并基于此模型改进了前撞预警系统报警策略。驾驶风格分类结合了分位点法和信息熵法，以不同方式进行特征提取，使用k?means方法聚类样本数据。基于长短期记忆模型，设计了编码器-解码器模型用于预测，以上述分类的全部数据训练模型的共用部分参数，以提高模型的泛化能力；而以3种驾驶风格对应数据集对模型的个性化部分参数进行更高学习率的训练。利用上述预测模型，对基于欧洲新车评估计划-自动紧急制动系统测试协议的前撞预警策略进行改进，使误报警次数从123降低至50。

关键词: 驾驶风格, 纵向相对距离预测, 前撞预警策略

Abstract:

With consideration of driving style, an optimized prediction model for longitudinal relative distance is designed, based on which an alarm strategy for the frontal?crash warning system is improved. The combination of quantile method and information entropy method is adopted for driving style classification to extract features in different ways, and k?means method is used to cluster sample data. Based on long short-term memory model, the encoder?decoder model is designed for prediction. All the data of above classifications are used to train the sharing parameters of model for improving its generalization ability, while the personalized parameters are trained with a higher learning rate by the corresponding data set of three driving styles. Utilizing the above?mentioned prediction models, the warning strategy for frontal crash based on European NCAP?AEB test protocol is improved, and as a result, the number of false alarms reduces from 123 to 50.

Key words: driving style, longitudinal relative distance prediction, frontal crash warning strategy

金辉,李昊天. 基于驾驶风格的前撞预警系统报警策略[J]. 汽车工程, 2021, 43(3): 405-413.

Hui Jin,Haotian Li. Alarm Strategy for Frontal Crash Warning System Based on Driving Style[J]. Automotive Engineering, 2021, 43(3): 405-413.

图/表 17

图 1

表1

表2

表3

表4

表5

表6

表7

图 2

图 3

图 4

图 5

图 6

表8

图 7

图 8

图 9

参考文献 24

1	SAGBERG F， SELPI， PICCININI G F， et al. A review of research on driving styles and road safety［J］. Human Factors， 2015， 57（7）： 1248-1275.
2	FAROOQ D， MOSLEM S， TUFAIL R F， et al. Analyzing the importance of driver behavior criteria related to road safety for different driving cultures［J］. International Journal of Environmental Research and Public Health， 2020， 17（6）.
3	DE ONA J， DE ONA R， EBOLI L， et al. How to identify the key factors that affect driver perception of accident risk. A comparison between Italian and Spanish driver behavior［J］. Accident Analysis & Prevention， 2014， 73（1）： 225-235.
4	WANG J， ZHANG X， KANG H， et al. Test method and risk factor definition of forward collision warning system［J］. IEEE Access， 2020： 47730-47740.
5	PAMPEL S M， JAMSON S， HIBBERD D， et al. ACC design for safety and fuel efficiency： the acceptance of safety margins when adopting different driving styles［J］. Cognition， Technology & Work， 2020， 22（2）： 335-342.
6	SUN Q， ZHANG H， LI Z， et al. ADAS acceptability improvement based on self⁃learning of individual driving characteristics： a case study of lane change warning system［J］. IEEE Access， 2019： 81370-81381.
7	WANG C， SUN Q， GUO Y， et al. Improving the user acceptability of advanced driver assistance systems based on different driving styles： a case study of lane change warning systems［J］. IEEE Transactions on Intelligent Transportation Systems， 2019： 1-13.
8	CHENG S， WANG C， ZHANG S， et al. Study on control strategy for personalised lane⁃change on highway［J］. The Journal of Engineering， 2018， 2018（16）： 1724-1730.
9	WANG W， ZHAO D， HAN W， et al. A learning⁃based approach for lane departure warning systems with a personalized driver model［J］. IEEE Transactions on Vehicular Technology， 2018， 67（10）： 9145-9157.
10	TAKADA S， HIRAOKA T， KAWAKAMI H. Effectiveness of forward obstacles collision warning system based on deceleration for collision avoidance［J］. IET Intelligent Transport Systems， 2014， 8（6）： 570-579.
11	CHEN T， LIU K， WANG Z， et al. Vehicle forward collision warning algorithm based on road friction［J］. Transportation Research Part D： Transport and Environment， 2019， 66： 49-57.
12	WANG X， CHEN M， ZHU M， et al. Development of a kinematic⁃based forward collision warning algorithm using an advanced driving simulator［J］. IEEE Transactions on Intelligent Transportation Systems， 2016， 17（9）： 2583-2591.
13	GUO Q， ZHAO Z， SHEN P， et al. Adaptive optimal control based on driving style recognition for plug⁃in hybrid electric vehicle［J］. Energy， 2019， 186： 115824.
14	MURPHEY Y L， MILTON R， KILIARIS L. Driver's style classification using jerk analysis［C］.2009 IEEE Workshop on Computational Intelligence in Vehicles and Vehicular Systems. IEEE， 2009： 23-28.
15	DORR D， GRABENGIESSER D， GAUTERIN F， et al. Online driving style recognition using fuzzy logic［C］. International Conference on Intelligent Transportation Systems， 2014： 1021-1026.
16	LIU Y， WANG J， ZHAO P， et al. Research on classification and recognition of driving styles based on feature engineering［J］. IEEE Access， 2019： 89245-89255.
17	KARGINOVA N， BYTTNER S， SVENSSON M. Data⁃driven methods for classification of driving styles in buses［C］. SAE Paper 2012-01-0744.
18	MARTINEZ C M， HEUCKE M， WANG F， et al. Driving style recognition for intelligent vehicle control and advanced driver assistance： a survey［J］. IEEE Transactions on Intelligent Transportation Systems， 2018， 19（3）： 666-676.
19	WANG W， XI J， CHONG A Y， et al. Driving style classification using a semisupervised support vector machine［J］. IEEE Transactions on Human-Machine Systems， 2017， 47（5）： 650-660.
20	WANG W， XI J， ZHAO D， et al. Driving style analysis using primitive driving patterns with bayesian nonparametric approaches［J］. IEEE Transactions on Intelligent Transportation Systems， 2019， 20（8）： 2986-2998.
21	XUE Q， WANG K， LU J J， et al. Rapid driving style recognition in car⁃following using machine learning and vehicle trajectory data［J］. Journal of Advanced Transportation， 2019： 1-11.
22	HOCHREITER S， SCHMIDHUBER J. Long short⁃term memory［J］. Neural Computation， 1997， 9（8）： 1735-1780.
23	Mathworks. Forward collision warning using sensor fusion［EB/OL］. ［2020-08-11］. .
24	IRANMANESH S M， MORADIPARI E， FALLAH Y P， et al. Robustness of cooperative forward collision warning systems to communication uncertainty［C］. IEEE Systems Conference， 2016： 1-7.

下分位点对应概率	纵向相对距离下分位点/m
0	4.032
0.25	18.944
0.75	42.080
1	99.968

下分位点对应概率	纵向相对速度下分位点/（m·s^-1）	本车加速度下分位点/（m·s^-2）
0	-15	-4.991
0.2	-1.2	-0.413
0.4	-0.4	-0.180
0.6	0.2	0.013
0.8	0.9	0.250
1	15	4.630

变量序号	第一部分数据的变量	第二部分数据的变量
1	0.330	0.220
2	0.560	0.555
3	0.110	0.225
4	0.159	0.170
5	0.223	0.206
6	0.251	0.210
7	0.219	0.218
8	0.148	0.196
9	0.254	0.222
10	0.235	0.222
11	0.145	0.162
12	0.225	0.255
13	0.141	0.139

主成分序号	主成分解释度/%
1	43.93
2	30.00
3	12.63

组别	3.6 s+1.2 s组	6 s+2 s组
Seq2seq模型/m	0.828	1.341
一类seq2seq模型/m	0.546	1.038
二类seq2seq模型/m	0.552	1.084
三类seq2seq模型/m	0.587	1.196