面向纵向自动驾驶的仿人驱动控制网络模型

doi:10.3901/JME.2023.18.251

机械工程学报 ›› 2023, Vol. 59 ›› Issue (18): 251-262.doi: 10.3901/JME.2023.18.251

扫码分享

面向纵向自动驾驶的仿人驱动控制网络模型

高镇海^1,2, 于桐^1,2, 孙天骏^1,2, 唐明弘^1,2, 高菲^1,2, 赵睿^1,2

1. 吉林大学汽车仿真与控制国家重点实验室长春 130022;
2. 吉林大学汽车工程学院长春 130022

收稿日期:2022-11-09 修回日期:2023-05-29 出版日期:2023-09-20 发布日期:2023-12-07
通讯作者: 孙天骏(通信作者),男,1990年出生,博士,副教授,硕士研究生导师。主要研究方向为自动驾驶汽车拟人智能优化决策与控制方法。E-mail:sun_tj@jlu.edu.cn
作者简介:高镇海,男,1973年出生,博士,教授,博士研究生导师。主要研究方向为智能驾驶与人机交互。E-mail:gaozh@jlu.edu.cn;于桐,男,1998年出生,博士研究生。主要研究方向为驾驶员自然驾驶行为机理与自动驾驶仿人算法开发。E-mail:tongyu22@mails.jlu.edu.cn
基金资助:
国家自然科学基金(52202495,52202494,51775236)、吉林大学研究生创新基金(451230411061)、吉林大学长沙汽车创新研究院自由探索(CAIRIZT20220106)和中央高校基本科研业务费专项资金(2022-JCXK-24)资助项目。

Humanoid-driven Control Network Model for Longitudinal Autonomous Driving

GAO Zhenhai^1,2, YU Tong^1,2, SUN Tianjun^1,2, TANG Minghong^1,2, GAO Fei^1,2, ZHAO Rui^1,2

1. State Key Laboratory of Automobile Simulation and Control, Jilin University, Changchun 130022;
2. College of Automotive Engineering, Jilin University, Changchun 130022

Received:2022-11-09 Revised:2023-05-29 Online:2023-09-20 Published:2023-12-07

摘要/Abstract

摘要： 自动驾驶车辆已经成为当前汽车行业的研究热点。运动控制算法在很大程度上决定自动驾驶汽车的安全性和驾乘人员的接受度。自动驾驶控制算法的精度日益提高，然而其与驾驶员控制风格的一致性仍然较低，这会降低驾乘人员的体验感与接受度。此外，现有方法仍存在大量需要人工标定的参数，导致算法部署时的工作量较大。针对以上问题，提出车辆纵向驱动控制仿人神经网络(Vehicle longitudinal drive control human-like neural network，LCN)，并以此为基础构建了仿人机理的控制模型，其控制风格与驾驶员的一致性较高且能够实现参数的自学习标定。LCN的设计基于对人类驾驶行为的分析，其将驾驶员的控制机理与数据驱动方法相结合，并通过独特的网络架构设计将驾驶员机理中的容差控制特性与时延响应特性融入LCN。试验结果表明，所提出模型的控制风格更接近人类，且能够实现对车辆动力学特性的自估计与参数的自标定。

关键词: 自动驾驶, 仿人机理, 自学习标定, 运动控制, 容差控制

Abstract: Autonomous vehicles have become a current research hotspot in the automotive industry. Motion control algorithms remarkably determine the safety and passenger acceptance of autonomous vehicles. The accuracy of autonomous driving control algorithms is increasing, yet their consistency with the driver's control style is still low, which can reduce the driver's experience and acceptance. In addition, existing methods still have a large number of parameters that need to be manually calibrated, which leads to a high workload in algorithm deployment. A vehicle longitudinal drive control human-like neural network(LCN) is proposed, and a control model with humanoid mechanism is constructed based on it, which has a highly homogeneous control style and is capable of self-learning calibration of parameters. The design of the LCN is based on the analysis of human driving behavior. It combines the driver's control mechanism with a data-driven approach, and incorporates tolerance control characteristics and time-delay response characteristics in the driver's mechanism in the network architecture. The experimental results show that the control style of this model is closer to that of humans, and it is able to achieve self-estimation of vehicle dynamics and self-calibration of parameters.

Key words: autonomous vehicles, humanoid mechanism, self-learning calibration, motion control, tolerance control

中图分类号:

U46
TP399

高镇海, 于桐, 孙天骏, 唐明弘, 高菲, 赵睿. 面向纵向自动驾驶的仿人驱动控制网络模型[J]. 机械工程学报, 2023, 59(18): 251-262.

GAO Zhenhai, YU Tong, SUN Tianjun, TANG Minghong, GAO Fei, ZHAO Rui. Humanoid-driven Control Network Model for Longitudinal Autonomous Driving[J]. Journal of Mechanical Engineering, 2023, 59(18): 251-262.

参考文献

[1] PADEN B,CAP M,YONG S,et al. A survey of motion planning and control techniques for self-driving urban vehicles[J]. IEEE Transactions on Intelligent Vehicles,2016,1(1):33-55.
[2] CHEN Yan,WANG Junmin. Adaptive vehicle speed control with input injections for longitudinal motion independent road frictional condition estimation[J]. IEEE Transactions on Vehicular Technology,2011,60(3):839-848.
[3] SOUZA F D,STERN R. Calibrating microscopic car-following models for adaptive cruise control vehicles:Multiobjective approach[J]. Journal of Transportation Engineering Part a Systems,2021,147(1):04020150.
[4] YAO Di,ULBRICHT P,TONUTTI S,et al. A novel approach for experimental identification of vehicle dynamic parameters[J]. Proceedings of the Institution of Mechanical Engineers,Part D:Journal of Automobile Engineering,2020,234(10-11):2634-2648.
[5] DIELS C,BOS J E,HOTTELART K,et al. Motion sickness in automated vehicles:The elephant in the room[J]. Road Vehicle Automation 3,2016:121-129.
[6] HUANG Chao,LÜ Chen,HANG Peng,et al. Toward safe and personalized autonomous driving:Decision-making and motion control with DPF and CDT techniques[J]. IEEE/ASME Transactions on Mechatronics,2021,26(2):611-620.
[7] GERDES J C,HEDRICK J K. Vehicle speed and spacing control via coordinated throttle and brake actuation-sciencedirect[J]. Control Engineering Practice,1997,5(11):1607-1614.
[8] SHAKOURI P,ORDYS A,LAILA D S,et al. Adaptive cruise control system:Comparing gain-scheduling PI and LQ controllers[J]. IFAC Proceedings,2011,44(1):12964-12969.
[9] ZHU Min,CHEN Huiyan,XIONG Guangming. A model predictive speed tracking control approach for autonomous ground vehicles[J]. Mechanical Systems & Signal Processing,2017,87:138-152.
[10] ATTIA R,ORJUELA R,BASSET M. Combined longitudinal and lateral control for automated vehicle guidance[J]. Vehicle System Dynamics,2014,52(2):261-279.
[11] ZHOU Xianjie,VENHOVENS P. Calibration of vehicle dynamics and stability control model[J]. International Journal of Vehicle Systems Modelling & Testing,2016,11(4):285.
[12] HODGSON G,BEST M C. A Parameter identifying a kalman filter observer for vehicle handling dynamics[J]. Proceedings of the Institution of Mechanical Engineers,Part D:Journal of Automobile Engineering,2006,220(8):1063-1072.
[13] HONG S,LEE C,BORRELLI F,et al. A novel approach for vehicle inertial parameter identification using a dual kalman filter[J]. IEEE Transactions on Intelligent Transportation Systems,2015,16(1):151-161.
[14] LI Wenfei,LI Huiyun,XU Kun,et al. Estimation of vehicle dynamic parameters based on the two-stage estimation method[J]. Sensors (Basel,Switzerland),2021,21(11):3711.
[15] WENZEL T A,BURNHAM K J,BLUNDELL M V,et al. Dual extended Kalman filter for vehicle state and parameter estimation[J]. Vehicle System Dynamics,2006,44(2):153-171.
[16] GAO Zhenhai,YAN Wei,LI Hongjian. Design of the time-gap-dependent robust headway control algorithm for ACC vehicles[J]. International Journal of Vehicle Design,2016,70(4):325.
[17] ZHU Fan,XU Xin,MA Lin,et al. Autonomous driving vehicle control auto-calibration system:An industry-level,data-driven and learning-based vehicle longitudinal dynamic calibrating algorithm[C]//2020 IEEE Intelligent Vehicles Symposium (IV),2020:391-397.
[18] XIAO Xuesu,BISWAS J,STONE P. Learning inverse kinodynamics for accurate high-speed off-road navigation on unstructured terrain[J]. IEEE Robotics and Automation Letters,2021,6(3):6054-6060.
[19] DUAN Benming,WANG Qingnian,ZENG Xiaohua,et al. Calibration methodology for energy management system of a plug-in hybrid electric vehicle[J]. Energy Conversion and Management,2017,136:240-248.
[20] GAO Zhenhai,SUN Tianjun,ZHOU Shuhui,et al. Vehicle automatic iterative learning control based on drivers' starting behavior[C]//2019 Chinese Automation Congress (CAC). 2019.
[21] TAHERIAN S,KUUTTI S,VISCA M,et al. Self-adaptive torque vectoring controller using reinforcement learning[C]//2021 IEEE International Intelligent Transportation Systems Conference,ITSC 2021,September 19,2021-September 22,2021,2021:172-179.
[22] CHIANG H-T L,HSU J,FISER M,et al. RL-RRT:Kinodynamic motion planning via learning reachability estimators from rl policies[J]. IEEE Robotics and Automation Letters,2019,4(4):4298-4305.
[23] ZHANG Yu,TINO P,LEONARDIS A,et al. A survey on neural network interpretability[J]. IEEE Transactions on Emerging Topics in Computational Intelligence,2021,5(5):726-742.
[24] SALVUCCI,DARIO D. Modeling driver behavior in a cognitive architecture[J]. Human Factors,2006,48(2):362-380.
[25] TSIMHONI O,LIU Yili. Modeling steering using the queueing network-model human processor (QN-MHP)[C]//Proceedings of the Human Factors and Ergonomics Society Annual Meeting,2003:1875-1879.
[26] LINDORFER M,MECKLENBRAUKER C F,OSTERMAYER G. Modeling the imperfect driver:Incorporating human factors in a microscopic traffic model[J]. IEEE Transactions on Intelligent Transportation Systems,2018,19(9):2856-2870.
[27] LIAO Yuan,LI S E,WANG Wenjun,et al. Detection of driver cognitive distraction:A comparison study of stop-controlled intersection and speed-limited highway[J]. IEEE Transactions on Intelligent Transportation Systems,2016,17(6):1628-1637.
[28] 李松,许源源. 政策议程、传播与注意力:基于心理视角的分析[J]. 湖南社会科学,2018(6):83-91. LI Song,XU Yuanyuan. Policy agenda,communication and attention:Based on psychological perspective analysis[J]. Social Sciences in Hunan,2018(6):83-91.
[29] NARANJO J E,GONZALEZ C,GARCIA R,et al. Lane-change fuzzy control in autonomous vehicles for the overtaking maneuver[J]. IEEE Transactions on Intelligent Transportation Systems,2008,9(3):438-450.
[30] PARK J,SANDBERG I W. Universal approximation using radial-basis-function networks[J]. Neural Comput,1991,3(2):246-257.
[31] JACKSON I. Convergence properties of radial basis functions[J]. Constructive Approximation,1988,4(1):243-264.
[32] FEICHTENHOFER C,FAN H,MALIK J,et al. SlowFast networks for video recognition[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV),2019.
[33] XIE Siyue,HU Haifeng,CHEN Yizhen. Facial expression recognition with two-branch disentangled generative adversarial network[J]. IEEE Transactions on Circuits and Systems for Video Technology,2021,31(6):26.
[34] ZANGENEH E,RAHMATI M,MOHSENZADEH Y. Low resolution face recognition using a two-branch deep convolutional neural network architecture[J]. Expert Systems with Applications,2020,139:112854.
[35] XIE Ju,XU Xing,WANG Feng,et al. Modeling adaptive preview time of driver model for intelligent vehicles based on deep learning[J]. Proceedings of the Institution of Mechanical Engineers,Part I:Journal of Systems and Control Engineering,2021.
[36] RAMANISHKA V,CHEN Y T,MISU T,et al. Toward driving scene understanding:A dataset for learning driver behavior and causal reasoning[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),2018.
[37] ALT H,GODAU M. Computing the Fréchet distance between two polygonal curves[J]. International Journal of Computational Geometry & Applications,1995,5(1):75-91.

面向纵向自动驾驶的仿人驱动控制网络模型

Humanoid-driven Control Network Model for Longitudinal Autonomous Driving

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张立元, 杨锦波, 李澳, 杨庆凯, 徐光魁. 张拉整体球形机器人构型设计与控制研究进展[J]. 机械工程学报, 2024, 60(5): 1-18.
[2]	陈正, 吕立彤, 王飞, 姚斌, 李冬明, 刘红光, 张国良. 基于容腔压力规划的非模式切换负载口独立电液系统运动控制[J]. 机械工程学报, 2024, 60(2): 302-312.
[3]	褚端峰, 刘鸿祥, 高博麟, 王金湘, 殷国栋. 车辆队列预测巡航控制研究综述[J]. 机械工程学报, 2024, 60(18): 218-246.
[4]	袁小庆, 吴涛, 原勋, 王文东. 基于GSO-RF意图识别算法的全身助力外骨骼控制方法研究[J]. 机械工程学报, 2024, 60(17): 91-101.
[5]	隗寒冰, 吴化腾, 徐进. 考虑驾驶员NMS特征的自动驾驶汽车人机共驾鲁棒横向控制[J]. 机械工程学报, 2024, 60(16): 280-290.
[6]	高镇海, 于桐, 孙天骏. 考虑社会性行为的自动驾驶运动规划研究综述[J]. 机械工程学报, 2024, 60(10): 112-128.
[7]	戢杨杰, 张馨雨, 杨紫茹, 周上航, 黄岩军, 曹建永, 熊璐, 余卓平. 多智能网联汽车轨迹规划：现状与展望[J]. 机械工程学报, 2024, 60(10): 129-146.
[8]	梁凯冲, 赵治国, 颜丹姝, 赵坤. 基于动态运动基元的车辆高速公路换道轨迹规划[J]. 机械工程学报, 2024, 60(10): 192-206.
[9]	周洪龙, 裴晓飞, 刘一平, 赵柯帆. 面向动态不确定场景的自动驾驶车辆时空耦合分层轨迹规划研究[J]. 机械工程学报, 2024, 60(10): 222-234.
[10]	曾迪, 郑玲, 李以农, 杨显通. 自动驾驶奖励函数贝叶斯逆强化学习方法[J]. 机械工程学报, 2024, 60(10): 245-260.
[11]	聂士达, 刘辉, 廖志昊, 谢雨佳, 项昌乐, 韩立金, 林思豪. 考虑复杂地形的越野环境无人车辆路径规划研究[J]. 机械工程学报, 2024, 60(10): 261-272.
[12]	杨硕, 李时珍, 赵中原, 黄小鹏, 黄岩军. 基于时序差分学习模型预测控制的一体化自动驾驶换道策略[J]. 机械工程学报, 2024, 60(10): 329-338.
[13]	张奇祥, 王金湘, 张伊晗, 张荣林, 靳立强, 殷国栋. 智能电动汽车线控制动关键技术与研究进展[J]. 机械工程学报, 2024, 60(10): 339-365.
[14]	张志勇, 黄大洋, 黄彩霞, 胡林, 杜荣华. TD3算法改进与自动驾驶汽车并道策略学习[J]. 机械工程学报, 2023, 59(8): 224-234.
[15]	黄维维, 张鑫泉, 朱利民. 基于重复控制的快速刀具伺服系统前馈补偿方法[J]. 机械工程学报, 2023, 59(21): 43-51.