Open AccessA survey of adaptive optimal control theory
Author(s) -
Xiaoxuan Pei,
Kewen Li,
Yongming Li
Publication year - 2022
Publication title -
mathematical biosciences and engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.451
H-Index - 45
eISSN - 1551-0018
pISSN - 1547-1063
DOI - 10.3934/mbe.2022561
Subject(s) - optimal control , reinforcement learning , field (mathematics) , computer science , artificial neural network , control (management) , dynamic programming , constraint (computer aided design) , mathematical optimization , process (computing) , artificial intelligence , engineering , algorithm , mathematics , mechanical engineering , pure mathematics , operating system
This paper makes a survey about the recent development of optimal control based on adaptive dynamic programming (ADP). First of all, based on DP algorithm and reinforcement learning (RL) algorithm, the origin and development of the optimization idea and its application in the control field are introduced. The second part introduces achievements in the optimal control direction, then we classify and summarize the research results of optimization method, constraint problem, structure design in control algorithm and practical engineering process based on optimal control. Finally, the possible future research topics are discussed. Through a comprehensive and complete investigation of its application in many existing fields, this survey fully demonstrates that the optimal control algorithms via ADP with critic-actor neural network (NN) structure, which also have a broad application prospect, and some developed optimal control design algorithms have been applied to practical engineering fields.