基于T-S模糊变权重MPC的智能车轨迹跟踪控制

Trajectory Tracking Control of an Intelligent Vehicle Based on T-S Fuzzy Variable Weight MPC

为了协调智能驾驶车辆的轨迹跟踪精确性和稳定性,提高控制算法对不同工况的自适应能力,提出基于Takagi-Sugeno模糊变权重模型预测控制(Takagi-Sugeno fuzzy model predictive control,T-S FMPC)的轨迹跟踪控制策略。以前轮转角为控制变量建立MPC控制,并以实时横向位移误差和横摆角误差为模糊输入,通过T-S模糊控制在线优化MPC目标函数权重,协调权重矩阵对轨迹跟踪精确性和稳定性的影响。基于Carsim建立分布式驱动电动汽车的整车动力学模型,基于Simulink建立控制策略,通过双移线工况仿真及实车试验,验证了所提控制策略的有效性。仿真结果表明,相比于传统MPC控制,所提出的T-S模糊变权重MPC控制可降低横向位移误差达62.24%,有效提高轨迹跟踪精度;并且可使前轮转角波动减小37.46%、横摆角误差减小84.19%,显著增强轨迹跟踪稳定性;试验结果表明,在20 km/h、沥青路面双移线工况下,横向位移误差在0.12 m以内,横摆角误差在1°以内,且前轮转角控制曲线平滑,说明所提算法具有良好的控制效果和实用性。

In order to coordinate the trajectory tracking accuracy and stability of intelligent driving vehicles and improve the self-adaptive capability of the control algorithm to different working conditions,a trajectory tracking control strategy based on T-S(Takagi-Sugeno)fuzzy variable weight model predictive control(MPC)is proposed.The MPC control is established with the front wheel steering angle as the control variable.The real-time lateral displacement deviation and yaw angle deviation are taken as the fuzzy inputs.The MPC objective function weights are optimized online by T-S fuzzy control to coordinate the influence of the weight matrix on the trajectory tracking accuracy and stability.A whole vehicle dynamic model of distributed drive electric vehicle is established based on Carsim,the control strategy is established based on Simulink,and the effectiveness of the proposed control strategy is verified through dynamics simulation real vehicle test on double-lane change working condition.Simulation results show that,compared with the traditional MPC control,the proposed T-S fuzzy variable weight MPC control can reduce the lateral displacement deviation by 62.24%and effectively improve the trajectory tracking accuracy.Moreover,it can reduce the front wheel steering angle fluctuation by 37.46%and the yaw angle deviation by 84.19%,which significantly enhances the trajectory tracking stability.The test results show that the lateral displacement deviation is within 0.12 m and the yaw angle deviation is within 1°when the vehicle makes double lane change at 20 km/h on asphalt pavement.Meanwhile,the curve of front wheel steering angle is smooth,which indicates that the proposed algorithm has good control effect and practicality.