运动控制
联轴节(管道)
车辆动力学
运动(物理)
控制(管理)
计算机科学
汽车工程
控制工程
工程类
人工智能
机器人
机械工程
作者
Guoying Chen,Jun Yao,Hongyu Hu,Zhenhai Gao,Min Hua
出处
期刊:IEEE Transactions on Vehicular Technology
[Institute of Electrical and Electronics Engineers]
日期:2024-02-26
卷期号:73 (8): 10899-10915
被引量:4
标识
DOI:10.1109/tvt.2024.3369923
摘要
Due to the over-actuated characteristics, the distri-buted vehicle has more potential for mobility and controllability than the traditional vehicle, making it more suitable for autonomous driving. Motion control of distributed vehicles still faces significant challenges of adjusting longitudinal speed while tracking the reference trajectory and ensuring stability under complex and changeable driving maneuvers. This paper proposes a hierarchical coupling motion control method for the distributed vehicle considering vehicle posture at the handling limit. In the motion-following layer, a coupling motion controller is presented based on variable-time steps discrete linear time-varying model predictive control (VTDLTV-MPC). To improve the predictive model accuracy, the linearization reference point is updated with the predicted state and control sequence of the last moment. While performing coupling motion control, the vehicle posture is coordinated in the heading deviation to enable the reference sideslip angle tracking. By coordinated control of motion and posture, the distributed vehicle can be handled outside the stability region of the traditional vehicle. In the executive layer, a tire parameter online adaptive module (TPOAM) is proposed to update tire parameters online, which does not rely on complex and multi-parameter tire models or look-up tables. Finally, the proposed controller is verified on the hardware-in-the-loop (HIL) platform. The results show that the proposed controller can adaptively adjust the longitudinal speed and control the distributed vehicle at the friction circle boundary to fully utilize the chassis performance, while tracking the reference posture to realize the decoupling of yaw and lateral motions.
科研通智能强力驱动
Strongly Powered by AbleSci AI