视觉预瞄式智能车辆纵横向协同控制研究

A Research on Visual Preview Longitudinal and Lateral Cooperative Control of Intelligent Vehicle

针对智能车辆路径跟踪控制的高度非线性及纵横向运动控制的耦合性问题,结合视觉预瞄模型及模型预测控制理论提出了一种预瞄式MPC路径跟踪控制新方法。该方法首先对非线性预测模型进行局部线性化,将MPC最优化求解问题转化为二次规划问题。在每个控制时域内将纵向车速及预瞄距离视为已知,运用指数模型对纵向车速进行描述,并结合道路曲率及实际车速设计了预瞄距离发生器。仿真和试验结果表明:相较于传统MPC跟踪控制方法,本文中提出的预瞄式MPC控制方法在不同车速下均能减小跟踪过程中的横向偏差和方向偏差,提高了跟踪精度,且此提升效果在高速状态下更为明显。

In view of high nonlinearity and the coupling problem of longitudinal and lateral motion control of intelligent vehicle path tracking control, a new preview MPC path tracking control method is proposed based on visual preview model and model predictive control theory. Firstly, the nonlinear prediction model is locally linearized and the MPC optimization problem is transformed into a quadratic programming problem. Then, in each control time domain, the vertical speed and the preview distance are regarded as known. The exponential model is used to describe the vertical speed, and the preview distance generator is designed based on the road curvature and the actual speed. The simulation and experimental results show that compared with the traditional MPC tracking control method, the preview MPC control method proposed in this paper reduces the lateral deviation and direction deviation in the tracking process at different vehicle speeds and improves the tracking accuracy, which is more obvious at high speed.