Optimal Control Problem for Linear System Based on Adaptive Dynamics Programming and Gradient Descent Method

The optimal control problem is investigated for linear system with unknown dynamics in this paper. For linear system, adaptive dynamic programming (ADP) techniques and gradient descent method are combined to obtain an approximated optimal controller. We design ADP algorithms to calculate the system...

Full description

Saved in:
Bibliographic Details
Published in2021 11th International Conference on Intelligent Control and Information Processing (ICICIP) pp. 428 - 432
Main Authors Wu, Yanzhi, Liu, Lu
Format Conference Proceeding
LanguageEnglish
Published IEEE 03.12.2021
Subjects
adaptive dynamic programming
Approximation algorithms
Dynamic programming
gradient descent method
Heuristic algorithms
Information processing
Linear systems
Optimal control
System dynamics
Online AccessGet full text
DOI10.1109/ICICIP53388.2021.9642163

Cover

More Information
Summary:The optimal control problem is investigated for linear system with unknown dynamics in this paper. For linear system, adaptive dynamic programming (ADP) techniques and gradient descent method are combined to obtain an approximated optimal controller. We design ADP algorithms to calculate the system matrices of the linear system. Based on these calculated system matrices, a gradient descent algorithm is utilized to approximate the optimal feedback control gain. Finally, a numerical example is included for illustration.
DOI:10.1109/ICICIP53388.2021.9642163