农机使用与维修Issue(6)：37-41,5.DOI:10.14031/j.cnki.njwx.2025.06.008
基于自适应动态规划的多智能体系统协同控制研究
Research on Cooperative Control of Multi-intelligent Body System Based on Adaptive Dynamic Planning
摘要
Abstract
Traditional Multi-Agent Systems(Multi-Agent Systems,MAS)control methods are difficult to cope with the uncertainty and dynamics in complex environments.Based on the theory of Adaptive Dynamic Planning(ADP),this study constructs an Adaptive Dynamic Planning(ADP)framework suitable for MAS,and combines with Reinforcement Learning(RL)technology to design a data-driven distributed control strategy.Next,a neural network is used to ap-proximate the policy function and value function to achieve online learning and optimal control of the intelligent body in an unknown environment.Compared with traditional control methods,the cooperative control strategy based on adaptive dynamic planning can autonomously learn the optimal strategy and achieve better control performance with less a priori knowledge.The research results can provide new theoretical and technical support for intelligent robots,unmanned sys-tems,and other distributed control scenarios.关键词
智能体系统/自适应动态规划/协同控制/强化学习/分布式控制Key words
intelligent body systems/adaptive dynamic planning/cooperative control/reinforcement learning/distribu-ted control分类
信息技术与安全科学引用本文复制引用
杨晓惠,王彩霞,尹晓琳..基于自适应动态规划的多智能体系统协同控制研究[J].农机使用与维修,2025,(6):37-41,5.基金项目
河北省教育厅科学研究项目(ZC2024118) （ZC2024118）
