Finite-horizon Q-learning for discrete-time zero-sum games with application to H & INFIN;$$ {H}_{\infty } $$ control

Liu, Mingxiang1,2; Cai, Qianqian1,2,4; Meng, Wei1,2; Li, Dandan1,2; Fu, Minyue3

2023

10.1002/asjc.3027

ASIAN JOURNAL OF CONTROL   影响因子和分区

1561-8625

1934-6093

In this paper, we investigate the optimal control strategies for model-free zero-sum games involving the H(infinity )control. The key contribution is the development of a Q-learning algorithm for linear quadratic games without knowing the system dynamics. The finite-horizon setting is more practical than the infinite-horizon setting, but it is difficult to solve the time-varying Riccati equation associated with the finite-horizon setting directly. The proposed algorithm is shown to solve the time-varying Riccati equation iteratively without the use of models, and numerical experiments on aircraft dynamics demonstrate the algorithm's efficiency.

finite-horizon H(infinity)control linear quadratic (LQ) control Q-learning zero-sum games

SCI ; EI

英语

其他

National Natural Science Foundation of China["U21A20476","U1911401","62273100"] ; Guangdong Basic and Applied Basic Research Foundation["2020A1515011505","2021A1515012554"]

Automation & Control Systems

Automation & Control Systems

WOS:000918012600001

WILEY

20230313401524

Game theory ; Iterative methods ; Learning algorithms ; Optimal control systems ; Reinforcement learning

Artificial Intelligence:723.4 ; Machine Learning:723.4.2 ; Control Systems:731.1 ; Calculus:921.2 ; Numerical Methods:921.6 ; Probability Theory:922.1

ENGINEERING

Web of Science

1.Guangdong Univ Technol, Sch Automat, Guangzhou, Peoples R China
2.Guangdong Prov Key Lab Intelligent Decis & Coopera, Guangzhou, Peoples R China
3.Southern Univ Sci & Technol, Dept Mech & Energy Engn, Shenzhen, Peoples R China
4.Guangdong Univ Technol, Sch Automat, Guangzhou 510006, Guangdong, Peoples R China

Liu, Mingxiang,Cai, Qianqian,Meng, Wei,et al. Finite-horizon Q-learning for discrete-time zero-sum games with application to H & INFIN;$$ {H}_{\infty } $$ control[J]. ASIAN JOURNAL OF CONTROL,2023,25(4):3160-3168.

Liu, Mingxiang,Cai, Qianqian,Meng, Wei,Li, Dandan,&Fu, Minyue.(2023).Finite-horizon Q-learning for discrete-time zero-sum games with application to H & INFIN;$$ {H}_{\infty } $$ control.ASIAN JOURNAL OF CONTROL,25(4),3160-3168.

Liu, Mingxiang,et al."Finite-horizon Q-learning for discrete-time zero-sum games with application to H & INFIN;$$ {H}_{\infty } $$ control".ASIAN JOURNAL OF CONTROL 25.4(2023):3160-3168.