• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2026, Vol. 62 ›› Issue (8): 365-381.doi: 10.3901/JME.260216

• 特邀专辑:汽车线控底盘 • 上一篇    下一篇

扫码分享

基于多智能体理论的分布式驱动车辆ESC-TVC鲁棒协同控制策略研究

张年华1, 张永康1, 陈继成2, 李朋涛3, 李焱4, 张辉1   

  1. 1. 北京航空航天大学交通科学与工程学院 北京 102206;
    2. 北京航空航天大学可靠性与系统工程学院 北京 100191;
    3. 宁德时代(上海)智能科技有限公司 上海 201306;
    4. 舜泰汽车有限公司 淄博 255400
  • 收稿日期:2025-07-07 修回日期:2025-12-12 出版日期:2026-04-20 发布日期:2026-06-12
  • 作者简介:张年华,男,1999年出生,博士研究生。主要研究方向为车辆动力学控制。E-mail:zhangnh@buaa.edu.cn;张永康,男,1995年出生,博士研究生。主要研究方向为分布式驱动车辆控制。E-mail:yk1995@buaa.edu.cn;陈继成,男,1989年出生,博士,副教授。博士研究生导师。主要研究方向为车辆动力学和控制E-mail:jichengc@buaa.edu.cn;李朋涛,男,1987年出生,硕士,工程师。主要研究方向为分布式驱动和控制。E-mail:lipt02@catl.com;李焱,男,1973年出生,博士,副教授。主要研究方向为无人驾驶技术、虚拟现实与人机交互技术。E-mail:18874830973@139.com;张辉,男,1984年出生,博士,教授,博士研究生导师。主要研究方向为柴油机后处理系统,车辆动力学和控制,机电一体化,鲁棒控制和滤波以及网络控制系统。E-mail:huizhang285@buaa.edu.cn
  • 基金资助:
    山东省重点研发计划资助项目(2023CXGC010111)。

Robust Cooperative Control Strategy for ESC and TVC of Distributed-driven Vehicles Based on Multi-agent System

ZHANG Nianhua1, ZHANG Yongkang1, CHEN Jicheng2, LI Pengtao3, LI Yan4, ZHANG Hui1   

  1. 1. School of Transportation Science and Engineering, Beihang University, Beijing 102206;
    2. School of Reliability and Systems Engineering, Beihang University, Beijing 100191;
    3. CATL (Shanghai) Intelligent Technology Co., Ltd, Shanghai, 201306;
    4. Suntae Automobile Co., Ltd, Zibo 255400
  • Received:2025-07-07 Revised:2025-12-12 Online:2026-04-20 Published:2026-06-12

摘要: 随着新能源汽车技术的高速发展,分布式驱动架构展现出巨大的性能潜力,推动了扭矩矢量控制(Torque vector control,TVC)、电子稳定性控制(Electronic stability control,ESC)等关键技术的研究。不同于传统车辆,分布式驱动车辆的核心优势,是同时具备高精度动力学控制能力与高效的稳定性调控能力。与之对应的,TVC精准调控动力性能以及ESC保障车身横摆稳定,迫切需要功能融合与协同调控,以响应分布式驱动车辆横摆操纵性与横向稳定性两大核心需求。针对前述功能协同需求,考虑了TVC与ESC的耦合控制架构,并基于多智能体理论进行控制系统建模;然后,基于H控制理论,提出了鲁棒协同控制器设计方法,保障了驾驶员转角输入、TVC差扭驱动、ESC差分制动共同作用下,车辆横摆控制系统的稳定性;其次,通过相平面稳定域分析优化了ESC介入机制,并引入动态衰减因子与权重函数实现了操纵性与稳定性的平衡,降低了极限工况下,传统协同控制策略以稳定域为界,进行0-1切换的保守性;最后,通过AVL VSM/Simulink联合仿真,验证了所提出的协同控制策略在双移线、阶跃转向、正弦迟滞等典型工况下的有效性。经过对比分析,所提出的多智能体协同控制策略在非极限工况下,显著减少了不必要的ESC介入,将横摆角速度增益、纵向车速都维持在接近TVC水平;在极限工况下,能够调用ESC稳定车身横摆,保障行车安全。

关键词: 电子稳定性控制, 扭矩矢量控制, 分布式驱动汽车, 鲁棒控制, 多智能体理论

Abstract: With the rapid advancement of new-energy vehicle technologies, distributed-driven architectures have emerged as a paramount enabler of performance breakthroughs, catalyzing intensive research into torque-vectoring control(TVC) and electronic stability control(ESC). Different from the traditional vehicles, distributed-driven vehicles are demonstrated as an advantageous transportation system with high-precision motion control ability and high reliability stabilization ability. Accordingly, the effectiveness of such advantages requires more efforts to cooperate the TVC and ESC. Motivated by this, this paper respectively formulates the control systems of TVC and ESC based on the multi-agent system theory. Subsequently, with the H control theory, a design method of a robust cooperative controller for TVC and ESC is proposed, where the driver input, TVC-induced moment and ESC-induced moment are taken into account. Furthermore, based on the stable region analysis, dynamic attenuation factors and weighting functions are introduced to reduce the conservatism of the existing methods, where the TVC and ESC are switched within enable and disable via the judgement of inside or outside of the stable region. Finally, co-simulations are carried out with AVL VSM and Simulink to validate the proposed cooperative strategy under representative maneuvers such as double-lane change, step-steer, and sinusoidal-with-delay. By comparison, under normal conditions, the proposed strategy suppresses superfluous ESC interventions, preserving yaw-rate gain and longitudinal velocity at TVC-dominant levels, whereas in extreme scenarios it recruits ESC to stabilize the lateral motion of vehicles, thereby safeguarding safety.

Key words: electronic stability control(ESC), torque vector control(TVC), distributed-driven vehicles, robust control, multi-agent system

中图分类号: