汽车工程 ›› 2021, Vol. 43 ›› Issue (2): 153-161.doi: 10.19562/j.chinasae.qcgc.2021.02.001

• •    下一篇

基于NMPC的智能汽车纵横向综合轨迹跟踪控制

陈龙1,邹凯2,蔡英凤1(),滕成龙2,孙晓强1,王海2   

  1. 1.江苏大学汽车工程研究院,镇江 212013
    2.江苏大学汽车与交通工程学院,镇江 212013
  • 收稿日期:2020-04-22 修回日期:2020-07-08 出版日期:2021-02-25 发布日期:2021-03-04
  • 通讯作者: 蔡英凤 E-mail:caicaixiao0304@126.com
  • 基金资助:
    国家重点研发计划(2018YFB0105000);国家自然科学基金(51875255);江苏省自然科学基金(BK20180100);江苏省六大人才高峰项目(2018?TD?GDZB?022);江苏省战略性新兴产业发展重大专项(苏发改高技发(2016)1094号);镇江市重点研发计划(GY2017006)

Longitudinal and Lateral Comprehensive Trajectory Tracking Control of Intelligent Vehicles Based on NMPC

Long Chen1,Kai Zou2,Yingfeng Cai1(),Chenglong Teng2,Xiaoqiang Sun1,Hai Wang2   

  1. 1.Automotive Engineering Research Institute,Jiangsu University,Zhenjiang 212013
    2.School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang 212013
  • Received:2020-04-22 Revised:2020-07-08 Online:2021-02-25 Published:2021-03-04
  • Contact: Yingfeng Cai E-mail:caicaixiao0304@126.com

摘要:

本文中针对大曲率转弯工况下,智能汽车纵横向动力学特性的耦合和动力学约束导致轨迹跟踪精度和稳定性下降的问题,提出一种基于非线性模型预测控制(NMPC)的纵横向综合轨迹跟踪控制方法,通过NMPC和障碍函数法(BM)的有效结合,提高了跟踪精度,改善了行驶稳定性。首先建立四轮驱动-前轮转向智能汽车动力学模型和轨迹跟踪模型,采用非线性模型预测控制计算出期望的纵向力、侧向力和横摆力矩;然后基于轮胎动力学模型建立带约束的非线性规划数学模型,利用障碍函数法求解出四轮轮胎力的最优分配,并最终实现四轮驱动智能汽车纵横向综合轨迹跟踪控制。最后进行Carsim和Simulink联合仿真,结果表明,与传统的预瞄PID控制相比,所提方法可在考虑纵横向动力学耦合的情况下明显改善跟踪精度和行驶稳定性。

关键词: 智能汽车, 轨迹跟踪, 非线性模型预测控制, 障碍函数法

Abstract:

Aiming at the lowering of the trajectory tracking accuracy and stability caused by the coupling of longitudinal and lateral dynamic characteristics and the dynamic constraints of intelligent vehicles under large?curvature turning conditions, a longitudinal and lateral comprehensive trajectory tracking control method based on nonlinear model predictive control (NMPC) is proposed in this paper. Through the effective combination of NMPC and barrier (function) method (BM), the tracking accuracy and driving stability are improved. Firstly, a dynamics model for a four?wheel drive and front wheel steering vehicle and its trajectory tracking model are established and the NMPC is adopted to calculate the desired longitudinal force, lateral force and yaw moment. Then a nonlinear programming mathematical model with constraints is constructed based on tire dynamics model and the BM is used to solve out the optimal distribution of the tire forces of four?wheels, and finally the longitudinal and lateral comprehensive trajectory tracking control for a four?wheel drive intelligent vehicle is achieved. In the end, a Carsim and Simulink joint simulation is conducted with a result showing that compared with the traditional preview PID control, the method proposed can significantly improve the tracking accuracy and driving stability with consideration of the coupling between longitudinal and lateral dynamics characteristics.

Key words: intelligent vehicles, trajectory tracking, nonlinear model predictive control, barrier method