Trajectory tracking control technology for autonomous vehicles based on dynamic Kalman polynomial networks
Article Text (Baidu Translation)
-
摘要: 为实现复杂动力学场景下自动驾驶车辆高可靠性和高精度轨迹跟踪控制, 提出一种基于动态卡尔曼多项式网络的预测控制框架, 该控制框架通过利用时间序列记忆能力来捕捉车辆的动态特性, 融合卡尔曼滤波的自适应调整, 实现对实时控制误差的优化校正; 借助多项式网络的非线性特征扩展, 进一步提升系统的建模能力; 该方法集动态感知、误差修正与非线性建模于一体, 显著增强系统对车辆动力学变化的适应性与精度。此外, 还引用基于横向偏差和航向偏差的反馈控制器, 旨在与预测模块协同工作, 从而实现更加精确的路径跟踪误差校正。CarSim与Simulink联合仿真验证表明: 预测-反馈控制器在低附路面和高速紧急换道极限工况下展现出明显的优势, 其中高速紧急换道工况下与ILQR和NMPC控制器相比, 路径跟踪的横向偏差均方根值分别减少了32.5%和38.4%;航向偏差的均方根值则分别减少了37.8%和40.0%。所提方法可为自动驾驶系统中高精度、高可靠性的轨迹跟踪控制提供一种创新且有效的解决方案。Abstract: To achieve high-reliability and high-precision trajectory tracking control for autonomous vehicles in complex dynamic scenarios, a predictive control framework based on a dynamic Kalman polynomial network was proposed. The dynamic characteristics of vehicles were captured by utilizing time-series memory capabilities, and the adaptive adjustment of Kalman filter was integrated to achieve the optimal correction of real-time control errors. By utilizing the nonlinear feature extension of polynomial networks, the modeling capability of the system was further enhanced. Dynamic perception, error correction, and nonlinear modeling were integrated into the method, significantly enhancing the adaptability and precision of the system to vehicle dynamics variations. Furthermore, a feedback controller based on lateral and heading deviations was incorporated to work synergistically with the predictive module, thereby achieving more accurate path tracking error correction. The co-simulation verification of CarSim and Simulink shows that the predictive-feedback controller exhibits obvious advantages under extreme conditions of low-adhesion surfaces and high-speed emergency lane changes. Specifically, under the condition of high-speed emergency lane changes, compared with ILQR and NMPC controllers, the root mean square value of the lateral deviation of path tracking is reduced by 32.5% and 38.4%, respectively; the root mean square value of heading deviation is reduced by 37.8% and 40.0%, respectively. The proposed method can provide an innovative and effective solution for the high-precision and high-reliability trajectory tracking control in autonomous driving systems.
-
图 3 动态卡尔曼多项式网络预测模块
Figure 3. Diagram of the dynamic Kalman polynomial network prediction module
表 1 车辆参数
Table 1. Vehicle parameters
参数 数值 车身质量m/kg 1 525 前轴距lf/m 1.111 后轴距lr/m 1.756 车身转动惯量Iz/(kg·m2) 2 315 轮胎滚动半径r/m 0.325 轮距b/m 1.55 
下载: 导出CSV
表 3 场景Ⅰ控制目标误差
Table 3. Errors of control targets in case Ⅰ
参数 方法 最大误差 均方根误差 横向偏差/m ILQR控制器 0.174 1 0.076 5 NMPC控制器 0.184 9 0.092 0 预测-反馈控制器 0.192 6 0.095 7 航向偏差/rad ILQR控制器 0.026 1 0.012 6 NMPC控制器 0.029 2 0.015 4 预测-反馈控制器 0.031 5 0.014 0
下载: 导出CSV
表 4 场景Ⅱ控制目标误差
Table 4. Errors of control targets in case Ⅱ
参数 方法 最大误差 均方根误差 横向偏差/m ILQR控制器 0.249 6 0.127 2 NMPC控制器 0.257 3 0.130 3 预测-反馈控制器 0.239 1 0.120 7 航向偏差/rad ILQR控制器 0.053 6 0.021 7 NMPC控制器 0.048 2 0.019 8 预测-反馈控制器 0.044 1 0.020 7
下载: 导出CSV
表 5 场景Ⅲ控制目标误差
Table 5. Errors of control targets in case Ⅲ
参数 方法 最大误差 均方根误差 横向偏差/m ILQR控制器 2.425 3 0.739 2 NMPC控制器 2.541 0 0.810 7 预测-反馈控制器 1.721 2 0.499 3 航向偏差/rad ILQR控制器 0.219 6 0.069 1 NMPC控制器 0.242 8 0.071 7 预测-反馈控制器 0.133 9 0.043 0
下载: 导出CSV
表 6 各控制器变体在场景Ⅲ下的性能误差对比
Table 6. Performance error comparison of controller variants in case Ⅲ
参数 方法 最大误差 均方根误差 横向偏差/m 固定权重多项式网络预测-反馈控制器 2.915 5 1.054 2 动态卡尔曼多项式网络预测器(无反馈) 2.653 0 0.881 5 动态卡尔曼多项式网络预测-反馈控制器(完整方案) 1.721 2 0.499 3 航向偏差/rad 固定权重多项式网络预测-反馈控制器 0.296 4 0.099 7 动态卡尔曼多项式网络预测器(无反馈) 0.261 0 0.082 3 动态卡尔曼多项式网络预测-反馈控制器(完整方案) 0.133 9 0.043 0
下载: 导出CSV
-
[1] JU Z Y, ZHANG H, LI X, et al. A survey on attack detection and resilience for connected and automated vehicles: From vehicle dynamics and control perspective[J]. IEEE Transactions on Intelligent Vehicles, 2022, 7(4): 815-837. doi: 10.1109/TIV.2022.3186897 [2] CHENG S, WANG Z, YANG B, et al. Convolutional neural network-based lane-change strategy via motion image representation for automated and connected vehicles[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(9): 12953-12964. doi: 10.1109/TNNLS.2023.3265662 [3] 赵祥模国家重点研发计划(2021YFB2501200)团队. 自动驾驶测试与评价技术研究进展[J]. 交通运输工程学报, 2023, 23(6): 10-77. doi: 10.19818/j.cnki.1671-1637.2023.06.002ZHAO Xiang-mo's team supported by the National Key Research and Development Program of China(2021YFB2501200). Research progress in testing and evaluation technologies for autonomous driving[J]. Journal of Traffic and Transportation Engineering, 2023, 23(6): 10-77. doi: 10.19818/j.cnki.1671-1637.2023.06.002 [4] 李杰, 张洛维, 王晓燕, 等. 基于视锥距离和自适应权重卡尔曼滤波的多传感器融合算法[J]. 中国公路学报, 2024, 37(3): 194-203.LI Jie, ZHANG Luo-wei, WANG Xiao-yan, et al. A multi-sensor fusion algorithm based on view-cone distance and adaptive weighted Kalman filter[J]. China Journal of Highway and Transport, 2024, 37(3): 194-203. [5] ZHAO Y Y, WANG L, YUN X Y, et al. Enhanced scene understanding and situation awareness for autonomous vehicles based on semantic segmentation[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2024, 54(11): 6537-6549. doi: 10.1109/TSMC.2024.3403859 [6] HE X K, YANG H L, HU Z X, et al. Robust lane change decision making for autonomous vehicles: An observation adversarial reinforcement learning approach[J]. IEEE Transactions on Intelligent Vehicles, 2023, 8(1): 184-193. doi: 10.1109/TIV.2022.3165178 [7] 李胜琴, 丁雪梅. 基于五次多项式的智能车辆轨迹规划[J]. 江苏大学学报(自然科学版), 2023, 44(4): 392-398.LI Sheng-qin, DING Xue-mei. Trajectory planning of intelligent vehicle based on quintic polynomial[J]. Journal of Jiangsu University (Natural Science Edition), 2023, 44(4): 392-398. [8] SHI X Y, WANG H, CHEN L, et al. A distributed driving six-wheel steering commercial vehicle chassis stability domain criterion for coordination of multiple subsystems[J]. IEEE Transactions on Vehicular Technology, 2024, 73(12): 18512-18526. doi: 10.1109/TVT.2024.3445597 [9] 赵轩, 王姝, 马建, 等. 分布式驱动电动汽车底盘集成控制技术综述[J]. 中国公路学报, 2023, 36(4): 221-248.ZHAO Xuan, WANG Shu, MA Jian, et al. Review of chassis integrated control technology for distributed drive electric vehicles[J]. China Journal of Highway and Transport, 2023, 36(4): 221-248. [10] 黄益绍, 庄迪. 基于干扰观测器与终端滑模的车辆纵向控制[J]. 江苏大学学报(自然科学版), 2024, 45(5): 513-520.HUANG Yi-shao, ZHUANG Di. Vehicle longitudinal control based on disturbance observer and terminal sliding mode[J]. Journal of Jiangsu University (Natural Science Edition), 2024, 45(5): 513-520. [11] 潘世举, 李永乐, 李子先, 等. 基于改进纯跟踪的智能车路径跟随方法[J]. 汽车工程, 2023, 45(1): 1-8, 19.PAN Shi-ju, LI Yong-le, LI Zi-xian, et al. Path following method of intelligent vehicles based on improved pure tracking[J]. Automotive Engineering, 2023, 45(1): 1-8, 19. [12] 白国星, 孟宇, 刘立, 等. 无人驾驶车辆路径跟踪控制研究现状[J]. 工程科学学报, 2021, 43(4): 475-485.BAI Guo-xing, MENG Yu, LIU Li, et al. Current status of path tracking control of unmanned driving vehicles[J]. Chinese Journal of Engineering, 2021, 43(4): 475-485. [13] MARINO R, SCALZI S, NETTO M. Nested PID steering control for lane keeping in autonomous vehicles[J]. Control Engineering Practice, 2011, 19(12): 1459-1467. doi: 10.1016/j.conengprac.2011.08.005 [14] CHU D F, LI H R, ZHAO C Y, et al. Trajectory tracking of autonomous vehicle based on model predictive control with PID feedback[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(2): 2239-2250. [15] HU C, WANG R R, YAN F J, et al. Output constraint control on path following of four-wheel independently actuated autonomous ground vehicles[J]. IEEE Transactions on Vehicular Technology, 2016, 65(6): 4033-4043. doi: 10.1109/TVT.2015.2472975 [16] JUNG H, JUNG D, CHOI S B. LQR control of an all-wheel drive vehicle considering variable input constraint[J]. IEEE Transactions on Control Systems Technology, 2022, 30(1): 85-96. doi: 10.1109/TCST.2021.3051233 [17] NAM H, CHOI W, AHN C. Model predictive control for evasive steering of an autonomous vehicle[J]. International Journal of Automotive Technology, 2019, 20(5): 1033-1042. doi: 10.1007/s12239-019-0097-5 [18] ZHAI L, WANG C P, HOU Y H, et al. MPC-based integrated control of trajectory tracking and handling stability for intelligent driving vehicle driven by four hub motor[J]. IEEE Transactions on Vehicular Technology, 2022, 71(3): 2668-2680. doi: 10.1109/TVT.2022.3140240 [19] SHI X Y, WANG H, CHEN L, et al. Robust path tracking control of distributed driving six-wheel steering commercial vehicle based on coupled active disturbance rejection[J]. IEEE Transactions on Vehicular Technology, 2023, 72(11): 13940-13952. [20] HANG P, CHEN X B. Integrated chassis control algorithm design for path tracking based on four-wheel steering and direct yaw-moment control[J]. Proceedings of the Institution of Mechanical Engineers, Part Ⅰ: Journal of Systems and Control Engineering, 2019, 233(6): 625-641. [21] SPIELBERG N A, BROWN M, KAPANIA N R, et al. Neural network vehicle models for high-performance automated driving[J]. Science Robotics, 2019, 4(28): eaaw1975. doi: 10.1126/scirobotics.aaw1975 [22] DA LIO M, BORTOLUZZI D, PAPINI G P R. Modelling longitudinal vehicle dynamics with neural networks[J]. Vehicle System Dynamics, 2020, 58(11): 1675-1693. doi: 10.1080/00423114.2019.1638947 [23] BECKERS T, COLOMBO L J, HIRCHE S, et al. Online learning-based trajectory tracking for underactuated vehicles with uncertain dynamics[J]. IEEE Control Systems Letters, 2021, 6: 2090-2095. [24] YU H X, DUAN J M, TAHERI S, et al. A model predictive control approach combined unscented Kalman filter vehicle state estimation in intelligent vehicle trajectory tracking[J]. Advances in Mechanical Engineering, 2015, 7(5): 1-14. [25] LI L, WANG T Q, XIA Y Q, et al. Trajectory tracking control for wheeled mobile robots based on nonlinear disturbance observer with extended Kalman filter[J]. Journal of the Franklin Institute, 2020, 357(13): 8491-8507. doi: 10.1016/j.jfranklin.2020.04.043 [26] ZHANG Q, BHATTARAIN, CHEN H, et al. Vehicle trajectory tracking using adaptive Kalman filter from roadside lidar[J]. Journal of Transportation Engineering, Part A: Systems, 2023, 149(6): 04023043. doi: 10.1061/JTEPBS.TEENG-7535 [27] 陈特, 陈龙, 徐兴, 等. 分布式驱动无人车路径跟踪与稳定性协调控制[J]. 汽车工程, 2019, 41(10): 1109-1116.CHEN Te, CHEN Long, XU Xing, et al. Integrated control of unmanneddistributed driven vehicles path tracking and stability[J]. Automotive Engineering, 2019, 41(10): 1109-1116. [28] SHI X Y, WANG H, CHEN L, et al. Hybrid trigger cooperative control of six-wheeled commercial vehicles with multiple sub-systems based on sub-regional linearization model[J]. Simulation Modelling Practice and Theory, 2024, 135: 102973. doi: 10.1016/j.simpat.2024.102973 [29] HOU R F, ZHAI L, SUN T M, et al. Steering stability control of a four in-wheel motor drive electric vehicle on a road with varying adhesion coefficient[J]. IEEE Access, 2019, 7: 32617-32627. doi: 10.1109/ACCESS.2019.2901058 [30] 马永杰, 程时升, 马芸婷, 等. 卷积神经网络及其在智能交通系统中的应用综述[J]. 交通运输工程学报, 2021, 21(4): 48-71. doi: 10.19818/j.cnki.1671-1637.2021.04.003MA Yong-jie, CHENG Shi-sheng, MA Yun-ting, et al. Review of convolutional neural network and its application in intelligent transportation system[J]. Journal of Traffic and Transportation Engineering, 2021, 21(4): 48-71. doi: 10.19818/j.cnki.1671-1637.2021.04.003 [31] 吴骁, 史文库, 陈志勇. 基于交互式多模型卡尔曼滤波的主动悬架控制[J]. 汽车工程, 2023, 45(7): 1200-1211, 1253.WU Xiao, SHI Wen-ku, CHEN Zhi-yong. Active suspension control based on interacting multiple model Kalman filter[J]. Automotive Engineering, 2023, 45(7): 1200-1211, 1253. [32] 胡敬宇, 汪, 严永俊, 等. 基于限定记忆随机加权扩展卡尔曼滤波的车辆状态估计[J]. 东南大学学报(自然科学版), 2022, 52(2): 387-393.HU Jing-yu, WANG Yan, YAN Yong-jun, et al. Vehicle state estimation based on limited memory random weighted extended Kalman filter[J]. Journal of Southeast University (Natural Science Edition), 2022, 52(2): 387-393. [33] KAPANIA N R, GERDES J C. Design of a feedback-feedforward steering controller for accurate path tracking and stability at the limits of handling[J]. Vehicle System Dynamics, 2015, 53(12): 1687-1704. doi: 10.1080/00423114.2015.1055279 -
图(29) / 表(6)
计量
- 文章访问数: 69
- HTML全文浏览量: 32
- PDF下载量: 11
- 被引次数: 0
