计算机与数字工程2026，Vol.54Issue(6)：1569-1574,6.DOI:10.3969/j.issn.1672-9722.2026.06.005
基于强化学习的轨迹跟踪控制算法研究
Research on Model Predictive Trajectory Tracking Control Algorithm Based on Reinforcement Learning
摘要
Abstract
A model predictive control algorithm based on reinforcement learning is proposed for the trajectory tracking control problem of intelligent vehicles.An intelligent body in reinforcement learning is used to replace the prediction model,predict the op-eration decision of the future moment,design the trajectory tracking control reward function,and train the intelligent body so that the control algorithm can track the target trajectory quickly and accurately.The control algorithm is built and simulated by Matlab/Simulink to track the target trajectory under low-speed and high-speed operating conditions,respectively.The results show that the model predictive controller based on reinforcement learning has better dynamic tracking characteristics,the control tracking error is smaller,and the vehicle control process is more stable.关键词
强化学习/模型预测控制/轨迹跟踪Key words
reinforcement learning/model predictive control/trajectory tracking分类
信息技术与安全科学引用本文复制引用
赵环宇,冯樱..基于强化学习的轨迹跟踪控制算法研究[J].计算机与数字工程,2026,54(6):1569-1574,6.基金项目
湖北省中央引导地方科技发展专项(编号:2019ZYYD019)资助. （编号:2019ZYYD019）
