.png){kind=icon}
.png){kind=icon}
Return
Data-driven control for stochastic linear-quadratic optimal problem with completely unknown dynamics
.png){kind=icon}
PRE
AI
DOI：10.1080/00207721.2025.2474137.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study presents an off-policy iteration approach for finding adaptive optimal control policies online for continuous-time stochastic linear systems with completely unknown system dynamics. The proposed approach employs the approximate/adaptive dynamic programming technique to iteratively solve the stochastic algebraic Riccati equation using the online information of state and input, without requiring the priori knowledge of the system matrices. In addition, all iterations can be conducted by repeatedly using the same state and input information in some fixed time intervals. Theoretical guarantees are given for the stability of the closed-loop system and the convergence of the algorithm. Finally, the application of the proposed algorithm for two examples validates its feasibility and effectiveness.
Keywords:
Stochastic optimal control
adaptive dynamic programming
stochastic algebra equation
continuous-time system
Journal
I
IF:
4.6
Papers:
1.1K
Citations:
7.3K