Scholarlink

<p>To address the conflict between trajectory tracking accuracy and yaw-roll stability control objectives for distributed-drive intelligent electric vehicles under high-speed emergency obstacle avoidance scenarios, a coordinated control method based on adaptive-weight model predictive control (MPC) is proposed. Firstly, a vehicle state estimator integrating an Unscented Kalman Filter (UKF) and a Radial Basis Function Neural Network (RBFNN) is designed to improve the estimation accuracy of critical stability-related parameters such as sideslip angle, roll angle, and roll angular velocity under complex conditions. Based on yaw stability phase-plane analysis and lateral load transfer ratio, yaw and roll stability evaluation indices are established, enabling adaptive adjustment of control weights for trajectory tracking accuracy, yaw stability, and roll stability. An optimized torque distribution strategy and an anti-roll moment distribution strategy for the active suspension system are further developed to enhance overall control performance. Finally, simulation and hardware-in-the-loop (HIL) experiments were conducted. The results show that, compared with the baseline AFS + DYC method, the proposed adaptive MPC-based coordinated control reduced the peak lateral acceleration by 28%, the maximum roll angle by 17%, and the yaw-rate RMS error by 21%, demonstrating improved trajectory tracking accuracy, yaw–roll stability, and overall vehicle safety performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive MPC-based coordinated control of trajectory tracking and yaw–roll stability for distributed-drive electric vehicles

  • 期刊论文

  • 发表日期：

    2026年3月30日

查看全文
  • Linfeng Zhao,
  • Zijun Chen,
  • Wenbin Shao,
  • Jinfang Hu,
  • Xinqing Liang

摘要

To address the conflict between trajectory tracking accuracy and yaw-roll stability control objectives for distributed-drive intelligent electric vehicles under high-speed emergency obstacle avoidance scenarios, a coordinated control method based on adaptive-weight model predictive control (MPC) is proposed. Firstly, a vehicle state estimator integrating an Unscented Kalman Filter (UKF) and a Radial Basis Function Neural Network (RBFNN) is designed to improve the estimation accuracy of critical stability-related parameters such as sideslip angle, roll angle, and roll angular velocity under complex conditions. Based on yaw stability phase-plane analysis and lateral load transfer ratio, yaw and roll stability evaluation indices are established, enabling adaptive adjustment of control weights for trajectory tracking accuracy, yaw stability, and roll stability. An optimized torque distribution strategy and an anti-roll moment distribution strategy for the active suspension system are further developed to enhance overall control performance. Finally, simulation and hardware-in-the-loop (HIL) experiments were conducted. The results show that, compared with the baseline AFS + DYC method, the proposed adaptive MPC-based coordinated control reduced the peak lateral acceleration by 28%, the maximum roll angle by 17%, and the yaw-rate RMS error by 21%, demonstrating improved trajectory tracking accuracy, yaw–roll stability, and overall vehicle safety performance.