Controller Design of Tracking WMR System Based on Deep Reinforcement Learning
Round 1
Reviewer 1 Report
Comments to the authors
Manuscript ID: electronics-1612654
Title: Design of a Fuzzy PID Controller for Path Following of a Wheeled Mobile Robot
1) List of keywords is limited. Please add two or more relevant terms. This would increase visibility of your article.
2) Provide a list of contribution(s) at the end of Section 1. Also, provide a brief outline of the manuscript.
3) Literature review is just too weak. There are multiple recent publications on the subject that could be used to improve the literature review. For instance, consider discussing the following publications:
[R1]. “Path Following of Wheeled Mobile Robots Using Online-Optimization-Based Guidance Vector Field”, IEEE/ASME Transactions on Mechatronics, 2021. [DOI: 10.1109/TMECH.2021.3077911]
[R2]. “Constrained Control of UAVs in Geofencing Applications”, in Proc. 26th Mediterranean Conference on Control and Automation, 2018. [DOI: 10.1109/MED.2018.8443035]
[R3]. “Trajectory Tracking of an Omni-Directional Wheeled Mobile Robot Using a Model Predictive Control Strategy”, applied sciences, 2018. [DOI: 10.3390/app8020231]
4) Figure 6 presents the membership function. How to determine the shape and values? It should be properly justified.
5) Table 1: First row is “NB” for all possible cases. Is it correct?
6) Table 4: The value of k_i is too small. This means that the integral path does not contribute to the control input; that is you only apply a PD controller law to control the system. Consider comparing your method for other values of k_i.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This is the review of the manuscript entitled „Design of a Fuzzy PID Controller for Path Following of a Wheeled Mobile Robot”.
The subject may attract interest to the readers. In general, this manuscript is well organized and written, with a comprehensive literature review, detailing the framework approach of the study, clearly stated methodology, and nicely presented findings. The manuscript provides sufficient background information regarding the topic proposed.
Suggestion:
The authors should highlight the usefulness and scientific contribution of their study.
The authors can highlight the usefulness of the study in its practical applicability.
The authors should include a clear and concise paragraph about their research limitations and future.
The conclusion section is missing some perspective related to the future research work, quantifying main research findings.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This study modulated the PID controller gain parameter, as based on fuzzy control for tracking the wheeled mobile robot platform, and used a normalized system to reduce the effect of the read errors, as resulted from the environment and sensor process. However, the experimental results show that, on a complex tracking site, the tracking stability of the fuzzy PID of this system is higher than that of conventional PID control by 12.6%, maximum overshoot is reduced by 30.1%, and the tracking time is shortened by 5.18%.
This manuscript talks about fuzzy PID controller for wheel mobile robots (WMR) whereas when you read the paper, the WMR is treated like any other simply system without nonholonomic constraints and dynamics. This major simplification makes the problem trivial; so why do we need a fancy fuzzy PID controller? Also, the novelty of the PID fuzzy controller is not discussed. While the experimental results are nice, whereas, the study is based on a linear system with a fancy PID controller that is bound to work.
Having these comments, the paper is not ready for publication based on technical contributions; it may be suitable in an automation type magazine article.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
No further comment.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper's contribution is of question. While the authors tried to convince that the work is under the AI area, I do not see any new theoretical results on fuzzy PID control. Numerous papers have highlighted the same fact. The experimental results are interesting whereas the contributions are not there.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
Authors have now introduced deep reinforcement learning in the results and discussion section. This is not connected well with the other parts. The DRL-based implementation would be better.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
