Optimal Fuzzy Controller Design for Autonomous Robot Path Tracking Using Population-Based Metaheuristics^†

Round 1

Reviewer 1 Report

In this paper, optimization of autonomous path tracking in robot is considered by means of genetic algorithm. Fuzzy logic controller is utilized as control algorithm and some performance metrics are used for having judgments.

This paper needs a major revision before publication. I have some serious doubts about the formulation of optimization problem and problem statements in this paper so the authors should clarify this point before further investigations and I need to evaluate the paper after major revision considering the following concerns:

• Regarding the fact that this paper is considered by authors as a paper in optimization field, it is not clear how the optimization procedure is performed and what is the complexity of the proposed technique. This point should be clarified properly.
• It is quite unclear what the novelty of this paper are. In this regard, a complete discussion should be prepared in the introduction section about the novelty and the main concerns of this paper.
• Why GA and not other recent powerful techniques?
• In the abstract, some results of the numerical study should be mentioned briefly.
• In the abstract, detailed information about the design examples should be provided.
• In the introduction, the literature review is poor. Some of the most cited references in the area of optimum structural design should be mentioned.
• Since many references are cited in the introduction section, but some major discussions about the results and challenges of these references should be provided in order to compare to the results of this manuscript.
• Since the results of the proposed procedure are not compared to the results of other algorithms, other well-known metaheuristics should be used for having a complete comparative investigation should be conducted by consideration of the results of other recently proposed metaheuristic algorithms: GWO, AOA, CGO, Aquila Optimizer and some other approaches.
• The optimization problem statement in this paper is not in an acceptable level for a journal paper.
• Some of the important characteristics of the considered design examples should be provided. The provided information is not enough for conducting any further investigations by other researchers.
• The conclusion of the paper lacks the main quantitative results of this manuscript. It should be noted that some numerical results of the paper should be mentioned properly in the conclusion section.
• Number of the Objective Function Evaluation (OFE) in the GA method is not provided.
• Statistical analysis including non-parametric test such as Wilcoxon is also required for performance evaluation of the GA and alternative approaches (See Comments 8 and 3.
• Improve the English level of the paper and correct the typos.

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors propose an optimization method that solves the problem of autonomous path tracking using a rear-wheel fuzzy logic controller. The main advantage of this approach is that it enables the design of controllers using rules that are linguistically familiar to human users. In addition, the authors propose a new technique that uses three different paths to validate the performance of each candidate configuration. It extends the authors' previous work where they add two more membership functions to the previous fuzzy model, intending to have a finer-grained adjustment. Experiments show that it is a better solution than a published control law; the proposed fuzzy controller has a better RMSE-measured performance and has limitations concerning undesired yaw oscillation.
Additionally, the authors also identify problems with the common practice of evaluating the performance of fuzzy controllers with a single problem case and performance metric, resulting in controllers that tend to be overtrained. The paper can be accepted after small improvements. There should be a separate section Conclusion which does not appear in the current manuscript. The language should also be improved. Please try more emphasize what is novel in this paper. Finally, please describe in more detail in figure 3 parameters how it was chosen. 

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 3 Report

he authors propose a very good paper that continues their very good results in fuzzy logic, fuzzy modeling, fuzzy control, nature-inspired optimization and robotics.

The combination of the above fields is done correctly and in an attractive manner in this paper.

The authors specify correctly how is this paper extended compared to their recent INFUS-2021 paper.

The introduction certainly provides sufficient background. The introduction also includes all relevant references.    

The research design is appropriate. The methods are adequately described.

The results are clearly presented. The conclusions are supported by the results.

Please try to better connect the parameters a, b, ... in Figure 2 to the text.

Could you develop the application in Matlab as well?

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

No further comments.