| 引用本文: | 郭洪艳,李光尧,刘俊,郭景征,谭中秋,吕颖.滚动优化下的对偶启发规划车辆路径跟踪控制[J].控制理论与应用,2025,42(9):1746~1756.[点击复制] |
| GUO Hong-yan,LI Guang-yao,LIU Jun,GUO Jing-zheng,TAN Zhong-qiu,LV Ying.Dual heuristic programming vehicle path tracking control via receding horizon[J].Control Theory & Applications,2025,42(9):1746~1756.[点击复制] |
|
| 滚动优化下的对偶启发规划车辆路径跟踪控制 |
| Dual heuristic programming vehicle path tracking control via receding horizon |
| 摘要点击 3513 全文点击 184 投稿时间：2023-08-28 修订日期：2025-02-27 |
| 查看全文 查看/发表评论 下载PDF阅读器 HTML |
| DOI编号 10.7641/CTA.2024.30579 |
| 2025,42(9):1746-1756 |
| 中文关键词 车辆路径跟踪 对偶启发式规划 模型预测控制 强化学习 |
| 英文关键词 vehicle path tracking dual heuristic programming model predictive control reinforcement learning |
| 基金项目 国家自然科学基金项目(62373163,62103162),国家资助博士后研究人员计划项目(GZC20230948)资助. |
|
| 中文摘要 |
| 为提高智能车辆的路径跟踪精度,降低高速、大曲率工况下车辆模型不确定性对跟踪性能的影响,本文提
出了一种基于滚动优化对偶启发式规划(RHDHP)的智能车辆路径跟踪控制策略.首先,结合魔术公式建立了可表
征侧向轮胎力非线性特性的车辆系统模型.其次,设计了滚动优化思想下对偶启发式规划(DHP)的最优控制方法.
该方法中的DHP结构确保了车辆非线性特性下的近似最优解,滚动优化的引入提高了车辆系统对环境变化的自适
应性. 同时,从理论上分析了RHDHP方法的收敛性以及闭环系统的稳定性.最后,通过仿真验证了所提方法的有效
性. |
| 英文摘要 |
| To improve the path tracking accuracy of intelligent vehicles and reduce the impact of vehicle model un
certainty under high speed and large curvature scenarios, this paper proposes an intelligent vehicle path tracking control
strategy based on receding horizon dual heuristic programming (RHDHP). Firstly, a vehicle system model, capable of char
acterizing non linear characteristics of lateral tire forces, is established by combining with the magic formula. Subsequently,
an optimal control method using dual heuristic programming (DHP) under the framework of receding horizon optimization
is designed. The DHP structure in this method ensures an approximate optimal solution for the vehicle under nonlinear
characteristics, while the introduction of receding horizon optimization enhances the adaptability of the vehicle system to
environmental changes. Additionally, the convergence of the RHDHP method and the stability of the closed loop system
are theoretically analyzed. Finally, the effectiveness of the proposed method is verified through simulations. |
|
|
|
|
|