考虑车辆稳定的避障路径自适应模型预测跟踪

Adaptive Model Prediction Tracking of Obstacle Avoidance Path Considering Vehicle Stability

为了提高车辆在稳定行驶前提下的避障路径跟踪控制精度,设计了自适应模型预测路径跟踪控制器和稳定控制器。根据车辆运动特性,建立了车辆在横向和横摆向共三个自由度的车辆动力学方程;基于路径跟踪控制需要,建立了预瞄跟踪控制模型。在避障路径跟踪方面,提出了预测时域随车速和路径曲率自适应变化方法;以提高避障路径跟踪精度和控制量增量最小为优化目标,将路径跟踪控制问题转化为二次型问题,实现了自适应优化求解和时域滚动控制。在车辆行驶稳定方面,从车辆状态的稳定性预判、稳定力矩的计算和分配等方面设计了稳定控制器。经仿真验证,这里的方法能够将车辆状态控制在相平面的双线法稳定区域内,实现了车辆行驶的稳定性和安全性;在避障路径跟踪方面,这里的方法的最大跟踪误差和绝对误差均值远小于传统模型预测控制、文献[14]模糊滑模控制,充分证明了这里的方法在避障路径跟踪控制中的高精度和有效性。

In order to improve the accuracy of obstacle avoidance path tracking control under the premise of stable driving,an adaptive model predictive path tracking controller and a stability controller are designed.According to the vehicle motion character-istics,the vehicle dynamics equations with three degrees of freedom in the lateral and yaw directions are established;Based on the need of path tracking control,a preview tracking control model is established.In the aspect of obstacle avoidance path tracking,an adaptive change method of prediction time domain with vehicle speed and path curvature is proposed;In order to improve the accu-racy of obstacle avoidance path tracking and minimize the increment of control quantity,the path tracking control problem is trans-formed into a quadratic problem,and the adaptive optimization solution and time-domain rolling control are realized.In the as-pect of vehicle driving stability,the stability controller is designed from the aspects of stability prediction of vehicle state,calcula-tion and distribution of stability torque.The simulation results show that this method can control the vehicle state in the stable re-gion of the double line method in the phase plane,and realize the stability and safety of vehicle driving;In the aspect of obstacle avoidance path tracking,the maximum tracking error and absolute error mean value of this method are much smaller than the tra-ditional model predictive control and fuzzy sliding mode control in reference[14],which fully proves the high accuracy and effective-ness of this method in obstacle avoidance path tracking control.