基于神经网络的ESC线控制动压力估计与控制

doi:10.3901/JME.2025.04.219

摘要/Abstract

摘要： 汽车电子稳定性控制(Electronic stability controller，ESC)作为车辆主动安全领域的关键技术之一，不仅能在车辆制动抱死、驱动滑转、转向侧滑等工况下保障车辆处于安全状态，而且支持高级驾驶辅助系统(Advanced driving assistance system，ADAS)的线控制动需求。传统ESC的线控制动功能往往需要加装轮缸压力传感器从而实现压力闭环控制，为了降低基于ESC的线控制动产业化成本，采用基于BP神经网络的ESC线控制动压力估计方法，通过分析ESC线控制动过程中的压力控制模式，利用BP神经网络在多维度数据处理上具有拟合能力强的优势，在不同控制模式下对ESC液压特性进行建模，拟合后的模型可在给定的ESC电磁阀与电机控制指令下进行轮缸压力估计。同时提出一种基于逻辑门限选择的PI反馈压力控制策略，利用压力估计结果进行反馈控制，提高线控制动的压力控制精度。在无轮缸压力传感器的基础上，对搭载ESC控制器与泵体的硬件在环台架进行测试。最终试验结果表明，ESC线控制动压力估计方法具有较高的精度，满足反馈控制需求，同时压力控制的精度在0.3 MPa以内，满足ADAS和高级自动驾驶的线控制动需求，为低成本线控制动方案提供一种新思路。

关键词: 电子稳定性控制, BP神经网络, 线控制动, 压力估计与控制

Abstract: Automotive electronic stability controller(ESC) is one of the key technologies in the field of vehicle active safety. It can not only ensure that the vehicle is in a safe state under conditions such as vehicle braking lock, driving slip, and steering sideslip, but also support control-by-wire demand of advanced driving assistance system(ADAS). Traditional ESC’s control by wire function often requires the installation of wheel cylinder pressure sensors to achieve pressure closed-loop control. In order to reduce the industrialization cost of brake-by-wire based ESC, the pressure estimation method of ESC based on BP neural network is adopted. By analyzing the pressure control mode in the process of ESC brake-by-wire, the hydraulic characteristics of ESC are modeled under different control modes by taking advantage of the strong fitting ability of BP neural network in multi-dimensional data processing, the fitted model can estimate the wheel cylinder pressure under the given ESC solenoid valve and motor control command. At the same time, a PI feedback pressure control strategy based on logical threshold selection is proposed, which uses the pressure estimation results for feedback control to improve the pressure control accuracy. Based on without wheel cylinder pressure sensor, the hardware-in-the-loop bench equipped with an ESC controller and a pump body is tested. The final experimental results show that the ESC brake-by-wire pressure estimation method has high accuracy and satisfies feedback control. At the same time, the accuracy of pressure control is within 0.3 MPa, which meets the demand for brake-by-wire of ADAS and advanced automatic driving, and provides a new idea for low-cost brake-by-wire solutions.

Key words: electronic stability controller, BP neural network, brake-by-wire, pressure estimation and control

中图分类号:

U461

邵东, 尹思维, 李亮, 王翔宇, 魏凌涛, 周道林. 基于神经网络的ESC线控制动压力估计与控制[J]. 机械工程学报, 2025, 61(4): 219-228.

SHAO Dong, YIN Siwei, LI Liang, WANG Xiangyu, WEI Lingtao, ZHOU Daolin. Estimation and Control of ESC Brake-by-wire Pressure Based on Neural Network[J]. Journal of Mechanical Engineering, 2025, 61(4): 219-228.

参考文献

[1] KOISAARI T，KARI T，VAHLBERG T，et al. Crash risk of ESC-fitted passenger cars[J]. Traffic Injury Prevention，2019，20(3):325-331.
[2] VAN ZANTEN A T. Bosch ESP system:5 Years of experience[J]. SAE Transactions，2000，109(7):428-436.
[3] 宗长富，潘钊，胡丹，等. 基于扩展卡尔曼滤波的信息融合技术在车辆状态估计中的应用[J]. 机械工程学报，2009，45(10):272-277. ZONG Changfu，PAN Zhao，HU Dan，et al. Information fusion algorithm for vehicle state estimation based on extended Kalman filtering[J]. Journal of Mechanical Engineering，2009，45(10):272-277.
[4] KIM D，MIN K，KIM H，et al. Vehicle sideslip angle estimation using deep ensemble-based adaptive Kalman filter[J]. Mechanical Systems and Signal Processing，2020，144:106862.
[5] LIU Y，JI X，YANG K，et al. Finite-time optimized robust control with adaptive state estimation algorithm for autonomous heavy vehicle[J]. Mechanical Systems and Signal Processing，2020，139:106616.
[6] 苑磊，何仁. 基于线性自抗扰控制的汽车ABS滑移率控制研究[J]. 汽车工程，2021，43(9):1367-1374. YUAN Lei，HE Ren. Research on ABS slip ratio control of vehicle based on linear active disturbance rejection control[J]. Automotive Engineering，2021，43(9):1367-1374.
[7] 李静，石求军，洪良，等. 基于车辆状态估计的商用车ESC神经网络滑膜控制[J]. 吉林大学学报(工学版)，2020，50(5):1545-1555. LI Jing，SHI Qiujun，HONG Liang，et al. Commercial vehicle ESC neural network sliding mode control based on vehicle state estimation[J]. Journal of Jilin University (Engineering and Technology Edition)，2020，50(5):1545-1555.
[8] TAGHAVIFAR H，HU C，TAGHAVIFAR L，et al. Optimal robust control of vehicle lateral stability using damped least-square backpropagation training of neural networks[J]. Neurocomputing，2020，384:256-267.
[9] 陈杰. 电子稳定控制系统控制策略及关键问题研究[D]. 北京:清华大学，2017. CHEN Jie. Research on control strategy and key problems of electronic stability control system[D]. Beijing:Tsinghua University，2017.
[10] 孟爱红. ESC液压执行机构压力精确控制研究[D]. 北京:清华大学，2014. MENG Aihong. Research on precise pressure control of ESC hydraulic actuator[D]. Beijing:Tsinghua University，2014.
[11] 刘佳熙，李升波，王建强，等. 车辆智能巡航控制纵向动力学参数快速辨识方法[J]. 农业机械学报，2010，41(10):6-10. LIU Jiaxi，LI Shengbo，WANG Jianqiang，et al. Fast identification method of vehicle longitudinal dynamic parameters for intelligent cruise control[J]. Transactions of the Chinese Society of Agricultural Machinery，2010，41(10):6-10.
[12] 陈国迎，赵选铭，文良浒，等. 基于线性自抗扰控制的自动驾驶车辆纵向加速度控制算法研究[J]. 汽车技术，2021(9):10-16. CHEN Guoying，ZHAO Xuanming，WEN Liangxu，et al. Research on longitudinal acceleration control algorithm of autonomous vehicles based on linear active disturbance rejection control[J]. Automoblie Technology，2021(9):10-16.
[13] WANG Yijing，CAO Shizhuo，YANG Hongjiu，et al. Model predictive longitudinal control for autonomous electric vehicles withtracking differentiator[J]. International Journal of Systems Science，2021，52(12):2564-2579.
[14] 林甲胜. 汽车电子稳定性系统轮缸压力估算模型仿真及实验研究[D]. 北京:中国地质大学，2019. LI Jiasheng. Simulation and experimental study on wheel cylinder pressure estimation model of automotive electronic stability system[D]. Beijing:China University of Geosciences，2019.
[15] YOU S H，HAHN J O，CHO Y M，et al. Modeling and control of a hydraulic unit for direct yaw moment control in an automobile[J]. Control Engineering Practice，2006，14(9):1011-1022.
[16] 李静，沙宏亮，王伯平，等. 基于液压制动轮缸压力估算的车辆电子稳定性程序控制算法[J]. 吉林大学学报(工学版)，2010(6):1478-1481. LI Jing，SHA Hongliang，WANG Boping，et al. Control algorithm for vehicle electronic stability program based on brake wheel cylinder hydraulic pressure estimation[J]. Journal of Jilin University (Engineering and Technology Edition)，2010(6):1478-1481.
[17] 刘伟，丁海涛，郭孔辉. 车辆稳定性电控系统液压调节器开环压力估计[J]. 吉林大学学报(工学版)，2012，42(5):1095-1099. LIU Wei，DING Haitao，GUO Konghui. Open-loop pressure estimation for hydraulic control unit of vehicle electronic stability control system[J]. Journal of Jilin University (Engineering and Technology Edition)，2012，42(5):1095-1099.
[18] 熊璐，杨兴，冷博，等. 无压力传感器下的电子液压制动系统轮缸液压力控制[J]. 同济大学学报，2020，48(8):1199-1207. XIONG Lu，YANG Xing，LENG Bo，et al. Wheel-cylinder hydraulic pressure control of integrated-electro-hydraulic brake system without pressure sensors[J]. Journal of Tongji University，2020，48(8):1199-1207.
[19] LOHNER H，DOMINKE P，CAO C T，et al. Method and device for determining the pressure in brake systems:U.S. Patent 6，446，490[P]. 2002-9-10.
[20] LI Q，BEYER K W，ZHENG Q. A model-based brake pressure estimation strategy for traction control system[C]//SAE 2001 World Congress. United States:SAE International，2001:3-6.
[21] VELARDOCCHIA M. A methodology to investigate the dynamic characteristics of ESP and EHB hydraulic units[C]//SAE World Congress & Exhibition. United States:SAE International，2006:1191-1203.
[22] IMAMURA M. Brake pressure estimating apparatus and method:U.S. Patent 7，066，559[P]. 2006-6-27.
[23] 祁雪乐，宋健，王会义，等. 基于AMESim的汽车ESP液压控制系统建模与分析[J]. 机床与液压，2005，8(15):115-116. QI Xuele，SONG Jian，WANG Huiyi，et al. Modeling and analysis of automotive ESP hydraulic control system based on AMESim[J]. Machine Tool & Hydraulics，2005，8(15):115-116.
[24] 李亮，宋健，韩宗奇，等. 用于电子稳定程序(ESP)在线控制的液压模型和反模型[J].机械工程学报，2008，44(2):139-144. LI Liang，SONG Jian，HAN Zongqi，et al. Hydraulic model and inverse model for electronic stability program online control system[J]. Journal of Mechanical Engineering，2008，44(2):139-144.
[25] 王伟玮. ESC液压执行单元的动态特性分析与综合仿真平台的建立[D].北京:清华大学，2011. WANG Weiwei. Dynamic characteristic analysis of ESC hydraulic actuator and establishment of comprehensive simulation platform[D]. Beijing:Tsinghua University，2011.
[26] WANG J，YIN C，ZHANG J. The development of test bench for vehicle handling and stability[J]. Drive System Technique，2004(6):14-16.
[27] WANG Kun，LI Shixin，ZHANG Hai. Application research of BP neural network optimization based on firefly algorithm[C]//Journal of Physics:Conference Series. Bristol:IOP Publishing，2021:032040.
[28] YAN Jian，PAN Zhifu，TAN Jing，et al. Assessment of water quality by firefly algorithm based on BP neural network model[J]. South-to-North Water Transfers and Water Science & Technology，2020，18(4):104-110.
[29] LUO Dayong. Study on prediction method of coal spontaneous combustion temperature based on BP neural network[J]. Coal Technology，2020，39(9):111-112.
[30] ZHONG Youwen，WU Xiaoling. Effects of cost-benefit analysis under back propagation neural network on financial benefit evaluation of investment projects[J]. PLOS One，2020，15(3):e0229739.