Review Reports

Round 1

Reviewer 1 Report

The present work addresses the implementation of a Type-2 Fuzzy-PID controller to a UAV Power Line Tracker. Simulation results are obtained through python with the ROS interface using the Gazebo simulator. The performance of the Type-2 Fuzzy-PID is compared with a type-1 PID and classical PID. Theoretically, the article has no relevant contribution, but from the application point of view, the presented results are interesting. After the review, the following comments arise:

1.- Some variables are shown, and no descriptions of these variables are given. Some variables are described after equations 3 and 4. The authors could consider including a nomenclature table.

2.- Consider using the same format for all the tables.

 

Author Response

Dear reviewer, thank you very much for your attention and time in reviewing our manuscript. In the PDF attached, you can find the author's reply.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper suggests using an Interval type-2 fuzzy logic-based PID controller to align and track transmission power lines with a UAV, allowing the controller to adapt to different input variable magnitudes and handle noisy data. Through simulations on the Gazebo platform, the proposed method showed better results than its T1-PID variant and the traditional Zigler-Nichols tuning technique for both lateral and angular errors. The type-2 fuzzy system also proved its capability to handle uncertainties in measurements during tests where the UAV followed power lines. The authors stressed the importance of evaluating the rule sets based on the proposed application and the methodology on various UAV models using the ROS framework in Python scripts. 

The authors provide a good motivation for the problem addressed, which is the inspection of transmission power lines with UAVs and explain why a new approach is needed to improve the precision and robustness of the control system.

Overall, I recommend this paper for publication, as it presents a novel and well-implemented approach for the inspection of transmission power lines with UAVs, which can have significant practical applications. However, I would suggest that the authors provide more details on the limitations of their methodology, including the assumptions made and the potential challenges that may arise in real-world scenarios. Additionally, it would be useful to include more discussion on the scalability and practicality of their approach, including the computational requirements and the potential for further optimization.

 

 

Author Response

Dear reviewer, thank you very much for your attention and time in reviewing our manuscript. In the PDF attached, you can find the author's reply.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper presents an interesting topic of inspecting a transmission power line using imagery through an UAV using a traditional PID controller to follow the transmission line. The tuning method for the PID gains are done using a Type 2 Fuzzy logic rather than traditional Zigler Nichols method.

A few suggestions and comments are provided below to make the paper stand out.

1. Please highlight the key outcome of the paper in the abstract. How is the proposed approach better than available methods.

2. Please alphabetical order for keywords.

3.  Line 37, what is 4, 8 times means? Is it 4x and 8x times?

4. Appreciate the way you highlighted the major contribution of the paper in line 94.

5.  What is FOU? Please elaborate where it is used as a first instance.

6. It would be highly desirable if a transfer function of the plant model is presented. Are the rotors individually controlled? Is the problem coupled between inputs? Looks like the system is MIMO. In that case how is the problem decoupled? 

7. Is there any relationship deduced between the membership function for the fuzzy logic to tune the gains with the desired system performance requirements? Like how is the membership function selected for a certain overshoot, rise time, damping etc? Is there a difference or similarity with zigler nichols method in terms selecting these gains?

8. Based on the error Z, the gains are adaptively changed by the fuzzy logic. Does this creates discontinuity in the controller performance?

9. What is the input to the controller other than the gains?

10. RMSE is typically a better choice for calculating accumulated errors.

11. In the results discussion section it would be nice to quantify how/why the proposed approach is better that traditional gain tuning methods. Like how much more or less error, percentage comparison, convergence performance etc. 

Author Response

The author's notes to the reviewer are attached

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Thank you for considering my comments and promptly addressing all in detail. I do not have any further comments/concerns.