基于混合神经网络的汽车运动状态估计

doi:10.19562/j.chinasae.qcgc.2022.10.007

Abstract

Abstract:

For the problem that the existing vehicle motion state estimation algorithm relies heavily on the accuracy of the dynamic model and the accuracy is difficult to guarantee under large slip angle， the paper proposes a vehicle motion state estimation algorithm based on the hybrid neural network （HNN）. By analyzing the basic dynamic characteristics of the vehicle itself， an hybrid neural network architecture suitable for vehicle motion state estimation is designed， and the deep learning estimation of vehicle motion state is realized. Based on the dataset composed of multi standard operating conditions and typical real vehicle test conditions， network training and test verification are carried out. The results show that compared with the traditional algorithm， the proposed HNN algorithm realizes estimation of vehicle motion state without dynamic vehicle model， improves estimation accuracy， and is robust to road adhesion coefficient change.

Key words: vehicle state estimation, deep learning, gated recurrent unit, multilayer perceptron, hybrid neural network

Zhenhai Gao,Wenhao Wen,Minghong Tang,Jian Zhang,Guoying Chen. Estimation of Vehicle Motion State Based on Hybrid Neural Network[J].Automotive Engineering, 2022, 44(10): 1527-1536.

Figures/Tables 28

采集量	符号	说明
转向盘转角/（°）	$δ$	±540
驱动电机转矩/（N·m）	$D a$	［50，350］
左前轮转速/（r·min^-1）	$ω f l$	-
右前轮转速/（r·min^-1）	$ω f r$	-
左后轮转速/（r·min^-1）	$ω r l$	-
右后轮转速/（r·min^-1）	$ω r r$	-
纵向加速度/g	$a x$	簧上质量质心处
侧向加速度/g	$a y$	簧上质量质心处

估计量	车速/（km·h^-1）	对比算法
估计量	车速/（km·h^-1）	EKF	DNN	GRU	HNN
$v x$	30	0.240 6	0.087 0	1.104 1	0.019 8
	70	0.230 3	0.341 4	0.749 7	0.021 2
	120	2.052 8	0.627 0	2.024 6	0.044 9
$v y$	30	0.196 9	0.352 9	0.741 2	0.110 0
	70	0.940 2	0.299 8	0.500 2	0.279 1
	120	3.192 5	1.226 6	2.709 2	1.179 6
$ω r$	30	1.891 7	0.810 8	2.108 6	0.565 8
	70	0.723 5	0.520 2	2.044 3	0.381 5
	120	2.917 4	1.709 6	2.434 5	1.504 1

估计量	路面	对比算法
估计量	路面	EKF	DNN	GRU	HNN
$v x$	0.3	1.060 7	0.504 8	1.437 8	0.030 6
	0.5	0.925 5	0.502 4	1.508 3	0.029 5
	0.85	1.320 7	0.505 5	1.562 1	0.037 4
	1.0	2.192 7	0.506 7	1.593 7	0.041 5
$v y$	0.3	1.518 6	1.127 0	1.755 1	1.064 0
	0.5	1.675 7	0.970 7	1.882 9	0.922 6
	0.85	2.673 2	0.734 0	2.050 8	0.720 4
	1.0	3.145 3	0.661 8	2.085 4	0.660 5
$ω r$	0.3	2.208 5	1.407 3	2.328 9	1.133 4
	0.5	2.206 7	1.193 3	2.192 5	1.037 8
	0.85	2.056 1	1.232 0	2.227 2	1.066 7
	1.0	2.023 8	1.317 4	2.240 7	1.147 2

References 21

1	GUO H Y， CAO D P， CHEN H， et al. Vehicle dynamic state estimation： state of the art schemes and perspectives ［J］. IEEE-CAA J Automatica Sin， 2018， 5（2）： 418-431.
2	PENDLETON S D， ANDERSEN H， DU X X， et al. Perception， planning， control， and coordination for autonomous vehicles ［J］. Machines， 2017， 5（1）： 54.
3	SINGH K B， ARAT M A， TAHERI S. Literature review and fundamental approaches for vehicle and tire state estimation［J］. Vehicle System Dynamics， 2019， 57（11）： 1643-1665.
4	GUO H Y， CHEN H， XU F， et al. Implementation of EKF for vehicle velocities estimation on FPGA ［J］. IEEE Transactions on Industrial Electronics， 2013， 60（9）： 3823-3835.
5	KATRINIOK A， ABEL D. Adaptive EKF-based vehicle state estimation with online assessment of local observability ［J］. IEEE Transactions on Control Systems Technology， 2016， 24（4）： 1368-1381.
6	LIU J， WANG Z， ZHANG L， et al. Sideslip angle estimation of ground vehicles： a comparative study ［J］. IET Control Theory & Applications， 2020， 14（20）： 3490-3505.
7	KIM D， KIM G， CHOI S， et al. An integrated deep ensemble-unscented Kalman filter for sideslip angle estimation with sensor filtering network ［J］. IEEE Access， 2021， 9： 149681-149689.
8	CHINDAMO D， LENZO B， GADOLA M. On the vehicle sideslip angle estimation： a literature review of methods， models， and innovations ［J］. Applied Sciences， 2018， 8（3）： 355.
9	周卫琪，齐翔，陈龙，等. 基于无迹卡尔曼滤波与遗传算法相结合的车辆状态估计［J］. 汽车工程， 2019， 41（2）： 198-205.
	ZHOU W Q， QI X， CHEN L， et al. Vehicle state estimation based on the combination of unscented Kalman filtering and genetic algorithm［J］. Automotive Engineering， 2019， 41（2）： 198-205.
10	SCHMIDHUBER J. Deep learning in neural networks： an overview ［J］. Neural Netw， 2015， 61： 85-117.
11	KUUTTI S， BOWDEN R， JIN Y， et al. A survey of deep learning applications to autonomous vehicle control ［J］. IEEE Transactions on ITS， 2021， 22（2）： 712-733.
12	ROKONUZZAMAN M， MOHAJER N， NAHAVANDI S， et al. Model predictive control with learned vehicle dynamics for autonomous vehicle path tracking ［J］. IEEE Access， 2021， 9： 128233-128249.
13	PAN Y， NIE X， LI Z， et al. Data-driven vehicle modeling of longitudinal dynamics based on a multibody model and deep neural networks ［J］. Measurement， 2021， 180： 109541.
14	CHINDAMO D, GADOLA M. Estimation of vehicle side-slip angle using an artificial neural network [C]. 2nd International Conference on Mechanical, Aeronautical and Automotive Engineering (ICMAA 2018). 2018.10.1051/matecconf/201816602001.
15	TONOLI A, AMATI N, GHOSH J. A deep learning based virtual sensor for vehicle sideslip angle estimation: experimental results [C]. SAE Paper 2018-01-1089.
16	SPIELBERG N A， BROWN M， KAPANIA N R， et al. Neural network vehicle models for high-performance automated driving ［J］. Sci Robot， 2019， 4（28）： 13.
17	LECUN Y， BENGIO Y， HINTON G. Deep learning ［J］. Nature， 2015， 521（7553）： 436-444.
18	SRINIVASAN S， SA I， ZYNER A， et al. End-to-end velocity estimation for autonomous racing ［J］. IEEE Robotics and Automation Letters， 2020， 5（4）： 6869-6875.
19	GONZALEZ L P， SANCHEZ S S， GARCIA-GUZMAN J， et al. Simultaneous estimation of vehicle roll and sideslip angles through a deep learning approach ［J］. Sensors， 2020， 20（13）： 18.
20	HOCHREITER S， SCHMIDHUBER J. Long short-term memory ［J］. Neural Computation， 1997， 9（8）： 1735-1780.
21	CHO K， VAN MERRI?NBOER B， GULCEHRE C， et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation［C］. Proceedings of the EMNLP 2014 - 2014 Conference on Empirical Methods in Natural Language Processing，Proceedings of the Conference， F， 2014.

[1]	LI Huan, HUANG Ying, ZHANG Fu-Jun, ZHAO Yu, GE Yan-Wu. [J]. , 2016, 38(12): 1420 -1426 .
[2]	HAO Han, WANG Si-南, LI Xiao, LIU Zong-Wei, ZHAO Fu-Quan. [J]. , 2017, 39(1): 1 -8 .
[3]	XU Cheng-Shan, JIANG Fa-Chao, SONG Sen-Nan, TIAN Guang-Yu. [J]. , 2017, 39(1): 9 -14 .
[4]	ZHANG Yi-Bing, 吕Jian-Jun , YUAN Guang-Hui. [J]. , 2016, 38(12): 1467 -1470 .
[5]	ZHANG Lei, ZHANG Jin-Qiu, LUO Tao, BI Zhan-Dong, YAO Jun. [J]. , 2016, 38(12): 1494 -1499 .
[6]	CUI An, LI Bin, WANG Xue-Liang, ZHANG Shi-Zhan. [J]. , 2016, 38(12): 1521 -1525 .
[7]	FENG Chong, ZHANG Dong-Hao, LUO Yu-Gong, LI Ke-Qiang. [J]. , 2017, 39(1): 66 -72 .
[8]	HAN Xiao-Mei, LIN Xue-Dong, LI De-Gang. [J]. , 2017, 39(1): 23 -27 .
[9]	YU Zhuo-Ping, HAN Wei, XIONG Lu. [J]. , 2017, 39(1): 52 -60 .
[10]	WEI Yin-Tao, ZHOU Fu-Qiang, LIU Zhe, XIANG Da-Bing. [J]. , 2017, 39(1): 92 -96 .

神经网络模型	GRU	DNN	DNN（包含动力系统状态输入）
RMSE	2.374 2	0.034 1	0.027 6
变化率		↓98.6%	↓19.1%
MAE	1.976 1	0.027 7	0.018 5
变化率		↓98.6%	↓33.2%

神经网络模型	DNN	GRU	GRU（输入包含纵向速度估计结果）
RMSE	2.214 5	1.250 8	1.099 9
变化率	-	↓43.5%	↓12.1%
MAE	1.897 8	0.982 9	0.747 1
变化率	-	↓48.2%	↓24.0%

本车转向操作	道路摩擦因数	车速/（km·h^-1）	转向盘转角/（°）
左侧双移线	0.3	20-70
左侧双移线	0.5，0.85，1.0	20-120
右侧双移线	0.3	20-70
右侧双移线	0.5，0.85，1.0	20-120
左侧J型转向	0.3，0.5， 0.85，1.0	20-120	40，60，90，100，120
右侧J型转向	0.3，0.5， 0.85，1.0	20-120	40，60，90，100，120
正弦转向0.2 Hz	0.5，0.85	30-60	±60，±90
正弦转向0.5 Hz	0.5，0.85	30-60	±60，±90

本车纵向操作	道路摩擦因数	加速踏板开度/％	加速踏板开度最大值/％
固定加速踏板开度	0.3，0.5，0.85，1.0	10-100
加速踏板开度平滑增大	0.3，0.5，0.85，1.0		30，60，90

项目	参数	取值
网络架构	隐层数	4
	第1层单元数	256
	第2层单元数	256
	第3层单元数	128
	第4层单元数	128
	参数个数	118 145
	激活函数	Relu
训练过程	小批量大小	32
	学习率初值	0.001
	提前停止耐心值	20
训练算法	梯度衰减因子	0.9
	平方梯度衰减因子	0.99
	偏置小量	1×10^-8

Estimation of Vehicle Motion State Based on Hybrid Neural Network

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 28

References 21

Related Articles 15

Metrics

Comments

Recommended 10

项目	参数	取值
网络架构	输入序列长度	5
	隐层数	1
	GRU单元数	200
	全连接隐层单元数	64
	dropout率	0.075
	参数个数	140 794
	激活函数	Relu
训练过程	小批量大小	32
	学习率初值	0.001
	提前停止耐心值	20
训练算法	梯度衰减因子	0.9
	平方梯度衰减因子	0.99
	偏置小量	1×10^-8

[1]	Ze Gao, Zunkang Chu, Jiasheng Shi, Fu Lin, Weixiong Rao, Haiyan Yu. Research on Fast Prediction Method of Stress Field of Automotive Parts Based on Graph Network [J]. Automotive Engineering, 2024, 46(1): 170-178.
[2]	Weiguo Liu,Zhiyu Xiang,Rui Liu,Guodong Li,Zixu Wang. Research on End-to-End Vehicle Motion Planning Method Based on Deep Learning [J]. Automotive Engineering, 2023, 45(8): 1343-1352.
[3]	Jiahao Zhao,Zhiquan Qi,Zhifeng Qi,Hao Wang,Lei He. Calculation of Heading Angle of Parallel Large Vehicle Based on Tire Feature Points [J]. Automotive Engineering, 2023, 45(6): 1031-1039.
[4]	Yanyan Chen,Hai Wang,Yingfeng Cai,Long Chen,Yicheng Li. Efficient Automatic Driving Instance Segmentation Method Based on Detection [J]. Automotive Engineering, 2023, 45(4): 541-550.
[5]	Fengchong Lan,Jikai Chen,Jiqing Chen,Xinping Jiang,Zihan Li,Wei Pan. Research on Lithium Battery Remaining Useful Life Prediction Method Driven by Real Vehicle Data [J]. Automotive Engineering, 2023, 45(2): 175-182.
[6]	Linhui Li,Xinliang Zhang,Yifan Fu,Jing Lian,Jiaxu Ma. Research on Visible Light and Infrared Post-Fusion Detection Based on TC-YOLOv7 Algorithm [J]. Automotive Engineering, 2023, 45(12): 2280-2290.
[7]	Xiaojun Zhang,Jingzhe Xi,Yanlei Shi,Anlu Yuan. Lightweight YOLOv7-R Algorithm for Road-Side View Target Detection [J]. Automotive Engineering, 2023, 45(10): 1833-1844.
[8]	Jie Hu,Yuanjie Li,Hao Geng,Huangzheng Geng,Xiong Guo,Hongwei Yi. Construction of Vehicle Fault Knowledge Graph Based on Deep Learning [J]. Automotive Engineering, 2023, 45(1): 52-60.
[9]	Jie Hu,Boyuan Xu,Zongquan Xiong,Minjie Chang,Di Guo,Lihao Xie. Cross-Domain Object Detection Algorithm Based on Multi-scale Mask Classification Domain Adaptive Network [J]. Automotive Engineering, 2022, 44(9): 1327-1338.
[10]	Guihong Bi,Xu Xie,Zilong Cai,Zhao Luo,Chenpeng Chen,Xin Zhao. Capacity Estimation of Lithium-ion Battery Based on Deep Learning Under Dynamic Conditions [J]. Automotive Engineering, 2022, 44(6): 868-878.
[11]	Dafang Wang,Jingdong Du,Jiang Cao,Mei Zhang,Gang Zhao. Research on Style Transfer Network for Autonomous Driving Data Generation [J]. Automotive Engineering, 2022, 44(5): 684-690.
[12]	Jinghua Guo,Zhifei He,Yugong Luo,Keqiang Li. Vehicle Cut-in Trajectory Prediction Based on Deep Learning in a Human-machine Mixed Driving Environment [J]. Automotive Engineering, 2022, 44(2): 153-160.
[13]	Yuande Jiang,Bing Zhu,Xiangmo Zhao,Jian Zhao,Bingbing Zheng. Modeling of Traffic Vehicle Interaction for Autonomous Vehicle Testing [J]. Automotive Engineering, 2022, 44(12): 1825-1833.
[14]	Datong Qin,Kang Wang,Jihao Feng,Yonggang Liu,Kun Cheng,Yu Xia. Road Slope Estimation Based on Unscented Kalman Filtering and Gated Recurrent Unit [J]. Automotive Engineering, 2022, 44(10): 1547-1555.
[15]	Shaowei Zhang,Dawei Zhu,Guangzhao Zhai. Prediction on Seat’s Anti-whiplash-injury Performance Based on Deep Learning [J]. Automotive Engineering, 2022, 44(10): 1600-1608.