变道过程非线性模型预测轨迹跟踪控制  

Nonlinear model predictive trajectory tracking control for lane changing process

作　　者：卢静[1] 金智林[2] LU Jing;JIN Zhilin(Jincheng College,Nanjing University of Aeronautics and Astronautics,Nanjing 200016,China;College of Energy and Power Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 200016,China)

机构地区：[1]南京航空航天大学金城学院,南京200016 [2]南京航空航天大学能源与动力学院,南京200016

出　　处：《重庆理工大学学报（自然科学）》2022年第10期58-65,共8页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金项目(51775269);汽车零部件先进制造技术教育部重点实验室开放课题(2016KLMT05);江苏省自然科学基金项目(BK20211190)。

摘　　要：针对变道过程中由于路面附着系数变化引起的模型非线性问题以及滚动优化算法对于实时性的要求,提出了变道过程非线性模型预测轨迹跟踪控制策略,分析了低路面附着系数引起的模型非线性特性,建立了变道过程轨迹跟踪模型及其前轮转角滚动时域优化求解方法。仿真分析及硬件在环测试可得,相比现有的线性轨迹跟踪方法,所提出的算法在高附着系数路面和低附着路面上的轨迹跟踪精度平均可提高14.35%,车辆横摆角速度跟踪误差平均可降低24.35%,侧向加速度误差平均可降低19.67%,有效实现了车辆在不同附着系数条件下的避障轨迹跟踪控制。Aiming at the model nonlinearity caused by the change of pavement adhesion coefficient in the process of lane change and the real-time requirements of rolling optimization algorithm,a nonlinear model predictive trajectory tracking control strategy in the process of lane change is proposed in this work.The model nonlinearity caused by low pavement adhesion coefficient is analyzed.Then,the track tracking model of lane changing process and the rolling time domain optimization method of front wheel angle are established.The simulation results and hardware in the loop test show that compared with the existing linear trajectory tracking methods,the trajectory tracking accuracy of the algorithm proposed in this paper can be improved by 14.35% on high adhesion road and low adhesion road,the vehicle yaw rate tracking error can be reduced by 24.35% and the lateral acceleration error can be reduced by 19.67%.Thus,the obstacle avoidance trajectory tracking control of vehicle under different adhesion coefficients is effectively realized.

关 键 词：无人驾驶汽车 轨迹跟踪 主动变道 路面附着系数 

分 类 号：U469.72[机械工程—车辆工程]