基于LTV-MPC的分布式驱动无人车路径跟踪与侧倾稳定性控制

doi:10.3969/j.issn.1674-8484.2026.03.012

摘要/Abstract

摘要：

针对分布式驱动无人车在高速变道或转向工况下横向载荷转移引发侧倾稳定性恶化问题，该文提出了一种考虑侧倾稳定性的分层协同路径跟踪控制策略。建立4自由度车辆模型，采用可变轮胎侧偏刚度，以提升高速载荷转移下模型精度；设计分层协同架构控制策略：上层基于线性时变模型预测控制（LTV-MPC），以路径跟踪性能与车辆稳定性为目标，将横向载荷转移率（LTR）引入优化约束；下层对各车轮驱动扭矩进行优化分配，提升轮胎利用率与车辆稳定性；最后，建立CarSim-Matlab/Simulink仿真平台，在高速双移线和鱼钩工况下进行验证。结果表明：相较于未改进的模型预测控制（MPC）策略，在100 km/h车速下，提出的改进策略在高速双移线工况下，LTR峰值和侧倾角峰值分别降低6.58%和11.92%；在鱼钩工况下，LTR峰值和侧倾角峰值分别降低10.64%和14.89%，避免了车辆失稳，验证了其在极限工况下的抗侧倾控制有效性。

关键词: 路径跟踪, 侧倾稳定性, 分层控制, 线性时变模型预测控制（LTV-MPC）, 横向载荷转移率（LTR）

Abstract:

A hierarchical cooperative path-tracking control strategy considering roll stability was proposed to address the deterioration of roll stability caused by lateral load transfer in distributed-drive autonomous vehicles during high-speed lane-change and steering maneuvers. Firstly, a four-degree-of-freedom vehicle model was established, and variable tire cornering stiffness was introduced to improve the model accuracy under high-speed lateral load transfer conditions. On this basis, a hierarchical cooperative control architecture was designed. In the upper layer, a linear time-varying model predictive control (LTV-MPC) method was developed to coordinate path-tracking performance and vehicle stability, where the lateral load transfer ratio (LTR) was incorporated as an explicit optimization constraint. In the lower layer, the driving torque of each wheel was optimally allocated to further improve tire utilization and vehicle stability. Finally, a CarSim-MATLAB/Simulink co-simulation platform was built to evaluate the proposed strategy under high-speed double lane change (DLC) and Fishhook maneuvers. The results show that, compared with the conventional MPC method, at a speed of 100 km/h, the proposed strategy reduces the peak LTR and peak roll angle by 6.58% and 11.92%, respectively, in the DLC maneuver; while in the Fishhook maneuver, reduces by 10.64% and 14.89%, respectively demonstrating that the proposed method can effectively suppress roll tendency and improve vehicle stability under extreme driving conditions.

Key words: path tracking, roll stability, hierarchical control, linear time-varying model predictive control (LTV-MPC), lateral load transfer ratio (LTR)

中图分类号:

U471.15

曹守启, 姜加胜, 周国峰, 陈渐伟. 基于LTV-MPC的分布式驱动无人车路径跟踪与侧倾稳定性控制[J]. 汽车安全与节能学报, 2026, 17(3): 397-408.

CAO Shouqi, JIANG Jiasheng, ZHOU Guofeng, CHEN Jianwei. Path tracking and roll stability control for distributed drive autonomous vehicles based on LTV-MPC[J]. Journal of Automotive Safety and Energy, 2026, 17(3): 397-408.

图/表 12

参考文献 19

[1]	ZHA Yunfei, DENG Jianxian, QIU Yinyuan, et al. A survey of intelligent driving vehicle trajectory tracking based on vehicle dynamics[J]. SAE Int’l J Vehi Dyna Stab NVH, 2023, 7(2): 221-248.
[2]	熊璐, 杨兴, 卓桂荣, 等. 无人驾驶车辆的运动控制发展现状综述[J]. 机械工程学报, 2020, 56(10): 127-143. doi: 10.3901/JME.2020.10.127
	XIONG Lu, YANG Xing, ZHUO Guirong, et al. A review of the development of motion control for autonomous vehicles[J]. J Mech Engi, 2020(10): 127-143. (in Chinese).
[3]	LI Lei, LI Jun, ZHANG Shiyi. Review article: state-of-the-art trajectory tracking of autonomous vehicles[J]. Mech Sci, 2021, 12(1): 419-432. doi: 10.5194/ms-12-419-2021
[4]	WANG Lihui, CHEN Zongliang, ZHU Wenxing. An improved pure pursuit path tracking control method based on heading error rate[J]. Indu Robot: Int’l J Robo Res Appl, 2022, 49(5): 973-980.
[5]	GU Jingming, FANG Dianjun. Genetic algorithm based LQR control for AGV path tracking problem[J]. J Phys Conf Seri IOP Publ, 2021, 1952(3): 12-32.
[6]	LU Ao, LU Ziwang, LI Runfeng, et al.Adaptive LQR path tracking control for 4WS electric vehicles based on genetic algorithm [C]// 2022 6th CAA Int’l Conf Vehi Cont Intel (CVCI). IEEE, 2022: 1-6.
[7]	ZHANG Lin, JIANG Lianbo, LIU Hanghang, et al. Cooperative strategy of trajectory tracking and stability control for 4WID autonomous vehicles under extreme conditions[J]. IEEE Trans Vehi Tech, 2023, 72(3): 3105-3118.
[8]	TIAN Yang, MA He, MA Lei, et al. Path tracking control of commercial vehicle emergency obstacle avoidance based on MPC and active disturbance rejection control[J]. IEEE Trans Transport Elect, 2025, 11(3): 7162-7170.
[9]	LIU Hebing, SUN Jinghong, CHENG Kawaieric. A two-layer model predictive path-tracking control with curvature adaptive method for high-speed autonomous driving[J]. IEEE Access, 2023, 11: 89228-89239. doi: 10.1109/ACCESS.2023.3306239 URL
[10]	CHEN Jie, WANG Ruochen, LIU Wei, et al. A review of recent advances in roll stability control in on-road and off-road vehicles[J]. Appl Sci, 2025, 15(10): 54-91.
[11]	Júnior Z R M, Murilo A, Lopes R V. Vehicle stability upper-level-controller based on parameterized model predictive control[J]. IEEE Access, 2022, 10: 21048-21065. doi: 10.1109/ACCESS.2022.3147452 URL
[12]	CHEN Shuping, XIONG Guangming, CHEN Huiyan, et al. MPC-based path tracking with PID speed control for high-speed autonomous vehicles considering time-optimal travel[J]. J Cent South Univ, 2020, 27(12): 3702-3720.
[13]	王康, 李琼琼, 王子洋, 等. 考虑侧倾的无人车NMPC轨迹跟踪控制[J]. 控制与决策, 2022, 37(10): 2535-2542.
	WANG Kang, LI Qiongqiong, WANG Ziyang, et al. Trajectory tracking control for automated vehicle based on NMPC considering vehicle rolling motion[J]. Contr Deci, 2022, 37(10): 2535-2542. (in Chinese).
[14]	江和耀, 王永海, 吴幼冬, 等. 四轮毂驱动电动车辆横向稳定与侧倾预防协同控制策略[J]. 吉林大学学报(工学版), 2024, 54(2): 540-549.
	JIANG Heyao, WANG Yonghai, WU Youdong, et al. Coordinated lateral stability and rollover prevention control for four-wheel independent motor drive electric vehicles[J]. J Jilin Univ (Engi Tech Edit), 2024, 54(2): 540-549. (in Chinese).
[15]	Ataei M, Khajepour A, Jeon S. Model predictive rollover prevention for steer-by-wire vehicles with a new rollover index[J]. Int’l J Contr, 2020, 93(1): 140-155.
[16]	Ghazali M, Durali M, Salarieh H. Vehicle trajectory challenge in predictive active steering rollover prevention[J]. Int’l J Autom Tech, 2017, 18(3): 511-521.
[17]	朱绍鹏, 吕超. 轮毂电机分布式驱动控制技术[M]. 北京: 机械工业出版社, 2022: 231.
	ZHU Shaopeng, LV Chao. In-Wheel Motor Distri-buted Drive Control Technology[M]. Beijing: China Machine Press, 2022: 231. (in Chinese).
[18]	Ataei M, Khajepour A, Jeon S. Development of a novel general reconfigurable vehicle dynamics model[J]. Mech Mach Theor, 2021, 156: 104-147.
[19]	WANG Zhaoqiang, SUN Keyang, MA Siqun, et al. Improved linear quadratic regulator lateral path tracking approach based on a real-time updated algorithm with fuzzy control and cosine similarity for autonomous vehicles[J]. Elect, 2022, 11(22): 3703.

整车质量，m / kg	1 860
簧上质量，m_s / kg	1 590
车辆轮距，W / m	1.575
质心到前轴的距离，a / m	1.18
质心到后轴的距离，b / m	1.77
车辆轴距，l / m	2.95
绕x轴的转动惯量，I_xx / (kg·m²)	894.4
绕z轴的转动惯量，I_zz / (kg·m²)	2 687.1
质心到侧倾中心的距离，h_s / m	0.57
侧倾中心到地面的距离，h_r / m	0.15
轮胎有效半径，R_m / m	0.393
侧倾刚度系数，k_φ / (Nm·rad^-1)	189 506
侧倾阻尼系数，c_φ / (Nm·s·rad^-1)	6 364

整车质量，m / kg	1 860
簧上质量，m_s / kg	1 590
车辆轮距，W / m	1.575
质心到前轴的距离，a / m	1.18
质心到后轴的距离，b / m	1.77
车辆轴距，l / m	2.95
绕x轴的转动惯量，I_xx / (kg·m²)	894.4
绕z轴的转动惯量，I_zz / (kg·m²)	2 687.1
质心到侧倾中心的距离，h_s / m	0.57
侧倾中心到地面的距离，h_r / m	0.15
轮胎有效半径，R_m / m	0.393
侧倾刚度系数，k_φ / (Nm·rad^-1)	189 506
侧倾阻尼系数，c_φ / (Nm·s·rad^-1)	6 364

方法	指标	e_y / m	e_? / rad	LTR	φ / rad	r / (rad·s^-1)
未改进 MPC策略	峰值	0.286 9	0.047 5	0.676 6	0.045 3	0.632 4
未改进 MPC策略	RMSE	0.083 2	0.014 2	0.292 9	0.018 8	0.059 6
LTV-MPC 策略	峰值	0.316 7	0.023 2	0.632 1	0.039 9	0.425 9
LTV-MPC 策略	RMSE	0.090 4	0.008 4	0.266 2	0.016 9	0.031 8

方法	指标	e_y / m	e_? / rad	LTR	φ / rad	r / (rad·s^-1)
未改进 MPC策略	峰值	0.286 9	0.047 5	0.676 6	0.045 3	0.632 4
未改进 MPC策略	RMSE	0.083 2	0.014 2	0.292 9	0.018 8	0.059 6
LTV-MPC 策略	峰值	0.316 7	0.023 2	0.632 1	0.039 9	0.425 9
LTV-MPC 策略	RMSE	0.090 4	0.008 4	0.266 2	0.016 9	0.031 8

方法	指标	e_y / m	e_? / rad	LTR	φ / rad	r / (rad·s^-1)
未改进MPC策略	峰值	4.691 3	0.404 0	0.758 3	0.049 7	0.527 4
未改进MPC策略	RMSE	1.871 8	0.107 6	0.399 7	0.025 5	0.219 7
LTV-MPC 策略	峰值	5.024 1	0.254 4	0.677 6	0.042 3	0.476 3
LTV-MPC 策略	RMSE	1.335 5	0.063 4	0.330 5	0.021 0	0.145 7