システム制御情報学会論文誌
Online ISSN : 2185-811X
Print ISSN : 1342-5668
ISSN-L : 1342-5668
論文
未知システムに対するワンショットな入出力履歴フィードバック制御器設計
—強化学習によるアプローチ
平井 卓実定本 知徳
著者情報
ジャーナル フリー

2021 年 34 巻 9 号 p. 235-242

詳細
抄録

In this paper, we propose a method of designing input-output history feedback controllers for unknown linear discrete-time systems. Many conventional reinforcement-learning based controls such as policy iteration are state-feedback. We extend the policy iteration by incorporating a method to statically estimate state variables from a history of finite-time input-output data. The convergence of the policy to model-based optimal solution has been theoretically guaranteed. Moreover, the proposed method is one-shot learning, i.e., the optimal controller can be obtained by using initial experiment data only. The effectiveness of the proposed method is shown through a numerical simulation through an oscillator network.

著者関連情報
© 2021 一般社団法人 システム制御情報学会
次の記事
feedback
Top