基于自适应MPC的拖拉机路径跟踪控制方法

Path Tracking Control Method of Tractor Based on Adaptive Model Prediction Control

针对传统模型预测控制(Model Predictive Control,MPC)无法适应拖拉机田间作业工况复杂性的问题,提出了一种基于自适应时域MPC的路径跟踪控制算法。首先,基于拖拉机运动学建立线性时变车辆误差模型;然后,根据拖拉机实时速度和参考路径曲率,利用模糊控制算法确定当前最优预测时域和控制时域;最后,结合MPC算法控制拖拉机跟踪预先设置好的参考路径。实车试验表明:拖拉机路径跟踪横向误差绝对值均值<0.03 m,直线段横向误差绝对值<0.07m,曲线段横向误差绝对值<0.15m,且路径跟踪控制器能够保证拖拉机在5s以内准确、快速地跟踪预先设置的参考路径。

In response to the problem that the traditional Model Predictive Control(MPC)cannot adapt to the complexity of tractor field operation conditions,an adaptive time-domain MPC-based path tracking control algorithm is proposed.First,a linear time-varying vehicle error model is established based on tractor kinematics.Then,based on the real-time speed of the tractor and the reference path curvature,the fuzzy control algorithm is used to determine the current optimal prediction time domain and control time domain.Finally,on this basis,the MPC algorithm is combined to control the tractor to track the pre-set reference path.The real vehicle test shows that:the absolute value of tractor path tracking lateral error is<0.03 m;the absolute value of straight section lateral error is<0.07m;the absolute value of curve section lateral error is<0.15 m;in addition,the path tracking controller designed in this paper can ensure that the tractor can accurately and quickly track the pre-set reference path within 5s.