Scholarlink

<p>Modeling and control of heavy-duty tractor-trailer systems are challenging due to system parameters variation and nonlinear dynamics. This paper proposes a hybrid physics-data-driven modeling and maneuver stability control framework for a tractor-trailer system. Specifically, an Integrated Physics-Data-Based (IPDB) approach models the highly nonlinear tractor-trailer system as a quasi Linear Parameter Varying (quasi-LPV) state-space model is proposed. The model integrates physical principles with system data snapshots, bypassing the requirement for accurate system parameters. Then, Model Predictive Control (MPC) based on the LPV model (i.e., LPV-MPC) is designed to address the constraints and achieve accurate tracking of maneuvering trajectory by previewing the vehicle dynamics in the prediction horizon. The constraint set and terminal cost are computed by Linear Matrix Inequality (LMI) that guarantees the closed-loop stability and recursive feasibility via Lyapunov analysis. To validate the proposed modeling and control strategy, Simulink-TruckSim co-simulation and hardware-in-the-loop experiment are conducted in snaking-turn and double-lane change scenarios. Across these scenarios, the proposed IPDB-MPC framework renders accurate modeling and tracking performance of yaw rates and lateral velocities of tractor and trailer.</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

Integrated physics-data-based modeling and LPV-MPC of heavy-duty tractor-trailer systems

  • 期刊论文

  • 发表日期：

    2026年7月28日

查看全文
  • Yidang Cai,
  • Liang Liu,
  • Yifan Men,
  • Tianyi He,
  • Wenpeng Wei

摘要

Modeling and control of heavy-duty tractor-trailer systems are challenging due to system parameters variation and nonlinear dynamics. This paper proposes a hybrid physics-data-driven modeling and maneuver stability control framework for a tractor-trailer system. Specifically, an Integrated Physics-Data-Based (IPDB) approach models the highly nonlinear tractor-trailer system as a quasi Linear Parameter Varying (quasi-LPV) state-space model is proposed. The model integrates physical principles with system data snapshots, bypassing the requirement for accurate system parameters. Then, Model Predictive Control (MPC) based on the LPV model (i.e., LPV-MPC) is designed to address the constraints and achieve accurate tracking of maneuvering trajectory by previewing the vehicle dynamics in the prediction horizon. The constraint set and terminal cost are computed by Linear Matrix Inequality (LMI) that guarantees the closed-loop stability and recursive feasibility via Lyapunov analysis. To validate the proposed modeling and control strategy, Simulink-TruckSim co-simulation and hardware-in-the-loop experiment are conducted in snaking-turn and double-lane change scenarios. Across these scenarios, the proposed IPDB-MPC framework renders accurate modeling and tracking performance of yaw rates and lateral velocities of tractor and trailer.