Convergence Analysis for an Online Data-Driven Feedback Control Algorithm

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper analyses the convergence of a numerical algorithm for stochastic optimal control problems.  The comments are as follows:

1, It might be a good idea to add more background and related literature on convergence analyses of numerical algorithms, which is the main problem studied in this paper.

2, It would be important to clarify the contributions of this paper relative to the previous work [2].

3, Most of the content in Section 2 has been presented in ref [2]. It would be better to shorten Section 2 considerably.

4, It is unclear to the reviewer why a data-driven control algorithm has been developed. What data has been collected and used? It seems to be just output feedback control. Please clarify.

5, It might be better to define notations in Section I.

6, What is the motivation for considering the specific probability metric in (27)? How is it related to the Wasserstein distance?

 

7, Moving lengthy proofs to the appendix might be better to improve readability. 

Comments on the Quality of English Language

n/a

Author Response

Please see our response in the attached pdf file. Thank you!

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In the reviewed paper, an interesting yet somewhat reduced dynamic problem was addressed. The effects of the convergence analysis of a feedback control algorithm based on online data were demonstrated. The work largely relied on existing theoretical assumptions and well-known methods. A linear problem was solved over a short period without discontinuities. Discontinuities in dynamic models of processes with a stochastic factor represent sudden changes in system behavior that are observed in reality and can result from various causes. These can be caused by abrupt changes in random variables affecting the systems, such as those studied in the paper. Therefore, two remarks are addressed:

I) Is it possible to test the presented methodology on a model with a jump discontinuity modeled by a Poisson process? These are important events.

II) How do the examined systems respond over a longer period, and are these solutions (in a closed-loop control system) still numerically stable?

 

Author Response

Please see our response in the attached pdf file. Thank you!

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

My doubts have been correctly addressed.