适用于路径跟踪控制自适应模型预测研究  

Research on adaptive model prediction for path tracking control

作　　者：李哲 刘飞 LI Zhe;LIU Fei(School of Mechanical and Automotive Engineering Shanghai University of Engineering Science,Shanghai 201620,China;School of Automotive Studies Tongji University,Shanghai 200092,China)

机构地区：[1]上海工程技术大学机械与汽车工程学院,上海201620 [2]同济大学汽车学院,上海200092

出　　处：《传感器与微系统》2025年第4期60-64,共5页Transducer and Microsystem Technologies

基　　金：河北省自然科学基金资助项目(E2016402066);河北省高层次人才项目(B2017003026)。

摘　　要：针对传统固定时域参数的模型预测控制(MPC)控制器很难满足车辆的稳定性和路径跟踪精度要求的问题,提出了一种基于时变速度选择最佳时域参数的自适应机制,通过分析控制器中时域参数对路径跟踪效果的影响,设计一种自适应时域参数选择器,将根据车速实时匹配到的最优时域参数输入到控制器中。在匹配到最佳时域参数的基础上,进一步考虑到车辆运动过程中跟踪精度与稳定性协调性问题,设计了基于模糊控制的权重系数自适应MPC控制器,模糊控制器将横向误差和航向误差作为输入,根据跟踪精度和稳定性的优先级对目标函数中权重进行修正。联合仿真结果表明:相较于传统控制器所提出的自适应路径跟踪控制器在跟踪精度与行驶稳定性方面均得到提升。Aiming at the problem that traditional model predictive controller with fixed time domain parameters is difficult to meet the requirements of vehicle stability and path tracking precision at the same time,an adaptive mechanism based on time-varying velocity to select optimal time domain parameters is proposed.By analyzing the impact of time domain parameters in the controller on path tracking effect,an adaptive time domain parameter selector is designed.The real-time matched optimal time domain parameters obtained according to vehicle velocity are then fed into the controller.Furthermore,on the basis of matching to the optimal time domain parameters,address the coordination between tracking precision and stability during vehicle motion,a weighted coefficient adaptive model predictive control(MPC)approach based on fuzzy control is designed.The fuzzy controller takes lateral error and heading error as inputs and according to the priority of tracking precision and stability,the weighting coefficients in the objective function is corrected.Joint simulation result demonstrates that the proposed adaptive path tracking controller enhances both tracking precision and driving stability compared to traditional controllers.

关 键 词：模型预测控制 自适应 预测时域 路径跟踪 模糊控制 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]