Stability Analysis of Model-Free Control under Iterative Q-learning Algorithms - 文章详情页

作者：

Zhao, Mingming (Zhao, Mingming.) | Wang, Ding (Wang, Ding.) | Qiao, Junfei (Qiao, Junfei.) | Gao, Ning (Gao, Ning.) | Xin, Peng (Xin, Peng.)

收录：

EI Scopus

摘要：

This　article　investigates　the　stability　of　the　closed-loop　system　under　the　iterative　control　policy　generated　by　various　iterative　Q-learning　algorithms.　First,　a　new　stability　criterion　is　developed　for　the　value-iteration-based　Q-learning　(VIQL)　algorithm,　which　is　initialized　by　a　positive　semi-definite　function.　Through　this　operation,　VIQL　can　provide　an　initial　admissible　control　policy　for　the　policy-iteration-based　Q-learning　(PIQL)　algorithm.　It　is　emphasized　that　evolving　control　policies　generated　by　PIQL　can　stabilize　the　controlled　system.　The　numerical　result　is　provided　to　verify　the　effectiveness　of　the　present　algorithms.　©　2023　IEEE.

关键词：

Learning algorithms Iterative methods Stability criteria Dynamic programming Reinforcement learning Closed loop systems

作者机构：

[ 1 ] [Zhao, Mingming]Beijing University of Technology, Faculty of Information Technology, Beijing, China
[ 2 ] [Wang, Ding]Beijing University of Technology, Faculty of Information Technology, Beijing, China
[ 3 ] [Qiao, Junfei]Beijing University of Technology, Faculty of Information Technology, Beijing, China
[ 4 ] [Gao, Ning]Beijing University of Technology, Faculty of Information Technology, Beijing, China
[ 5 ] [Xin, Peng]Beijing University of Technology, Faculty of Information Technology, Beijing, China

通讯作者信息：

电子邮件地址：

查看成果更多字段

成果类型
所属机构

所有年份指定年份从至