首页|Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
  • 维普
Reinforcement learning(RL)has roots in dynamic programming and it is called adaptive/approximate dynamic pro-gramming(ADP)within the control community.This paper reviews recent developments in ADP along with RL and its appli-cations to various advanced control fields.First,the background of the development of ADP is described,emphasizing the signifi-cance of regulation and tracking control problems.Some effec-tive offline and online algorithms for ADP/adaptive critic control are displayed,where the main results towards discrete-time sys-tems and continuous-time systems are surveyed,respectively.Then,the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed,respectively,where event-based design,robust stabi-lization,and game design are reviewed.Moreover,the extensions of ADP for addressing control problems under complex environ-ment attract enormous attention.The ADP architecture is revis-ited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation significantly.Finally,several typical control applications with respect to RL and ADP are summarized,particularly in the fields of wastewa-ter treatment processes and power systems,followed by some general prospects for future research.Overall,the comprehensive survey on ADP and RL for advanced control applications has demonstrated its remarkable potential within the artificial intelli-gence era.In addition,it also plays a vital role in promoting envi-ronmental protection and industrial intelligence.

Adaptive dynamic programming(ADP)advanced controlcomplex environmentdata-driven controlevent-triggered designintelligent controlneural networksnonlinear systemsopti-mal controlreinforcement learning(RL)

Ding Wang、Ning Gao、Derong Liu、Jinna Li、Frank L.Lewis

展开 >

Faculty of Information Technology,Beijing Key Laboratory of Computational Intelligence and Intelligent System,Beijing Laboratory of Smart Environmental Protection,and Beijing Institute of Artificial Intelligence,Beijing University of Technology,Beijing 100124,China

School of System Design and Intelligent Manufacturing,Southern University of Science and Technology,Shenzhen 518055,China,and also with the Department of Electrical and Computer Engineering,University of Illinois at Chicago,Chicago IL 60607 USA

School of Information and Control Engineering,Liaoning Petrochemical University,Fushun 113001,China

UTA Research Institute,the University of Texas at Arlington,Arlington TX 76118 USA

展开 >

国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金国家重点研发计划国家重点研发计划国家重点研发计划北京市自然科学基金

62222301620730856207315861890930-5620210032021ZD01123022021 ZD 01123012018YFC1900800-5JQ19013

2024

10.1109/JAS.2023.123843

自动化学报(英文版)
中国自动化学会,中国科学院自动化研究所,中国科技出版传媒股份有限公司

自动化学报(英文版)

CSTPCDEI
ISSN：2329-9266
年,卷(期)：2024.11(1)
  • 5