Scholarlink

<p>In this article, an autonomous stable drift control strategy for rear-wheel-drive vehicles is developed based on a linear time-varying model predictive control (LTV-MPC) algorithm. Considering the nonlinearity of the Fiala tire force model employed for drift control, a 3D lookup table for lateral force and lateral force gradient is designed to linearize the Fiala tire force model. Simultaneously, the controller transforms the autonomous drift control problem into a constrained optimal control problem using a three-degree-of-freedom (3-DOF) monorail vehicle model. Finally, the designed LTV-MPC algorithm is compared with a nonlinear model predictive control (NMPC) approach to evaluate control performance. Simulation results demonstrate that the controller achieves an 8.4 times improvement in computational speed compared to the NMPC algorithm while delivering comparable control performance, making it more suitable for deployment on real vehicles.</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

Autonomous Drift Control of Rear-Wheel Drive Vehicles Using Linear Time-Varying Model Predictive Control with Tire Model Linearization

  • 期刊论文

  • 发表日期：

    2026年3月5日

查看全文
  • Guangxin Wu,
  • Shaosong Li,
  • Liyuan Tian,
  • Xiaolei Pei,
  • Zaixiao Wang,
  • Gaojian Cui,
  • Xiaohui Lu

摘要

In this article, an autonomous stable drift control strategy for rear-wheel-drive vehicles is developed based on a linear time-varying model predictive control (LTV-MPC) algorithm. Considering the nonlinearity of the Fiala tire force model employed for drift control, a 3D lookup table for lateral force and lateral force gradient is designed to linearize the Fiala tire force model. Simultaneously, the controller transforms the autonomous drift control problem into a constrained optimal control problem using a three-degree-of-freedom (3-DOF) monorail vehicle model. Finally, the designed LTV-MPC algorithm is compared with a nonlinear model predictive control (NMPC) approach to evaluate control performance. Simulation results demonstrate that the controller achieves an 8.4 times improvement in computational speed compared to the NMPC algorithm while delivering comparable control performance, making it more suitable for deployment on real vehicles.