Motion Planning of UAV for Port Inspection Based on Extended RRT* Algorithm
Boris Galkin
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper presents an improved version of the RRT* algorithm for trajectory planning. It is based on the bias_RRT* algorithm and shortcuts from the sub-optimal trajectory using isoceles triangles.
The abstract is hard to understand.
"The goal of motion planning is to quickly generate an optimization trajectory that meets constraints. The solution of the paper is as follows...." Even though it isn't clarified, I assume that papers tries to give a solution to motion planning. Improve the text to make it clear. Also verify that the apropiate words are used, "optimization trajectory" or "optimal trajectory"
"apply a collision detection function", this function is not explained in the paper.
"Finally, the extended bias_RRT* algorithm is applied to two experimental setups." This is part of the solution? Please, make a clear distinction between the algorithm (obstacle detection, bias_RRT*, triangles, trajectory generation) and experiments
"The results show the path length and computation time are optimized, and the cost and deviation of trajectory are also optimized." None of the variables are optimized, because you don't find optimal values, they are improved values.
The organization of tha paper is Ok.
Section 2. "The purpose of path planning is to quickly generate a collision-free path without considering motion constraints."...."The bias_RRT* algorithm [14] has been used to generate a path with a simple greedy heuristic."
Why compare a path planning with a trajectory planning algorithm? it is unfair.
In the Simulation section, scenario 1: "In this scenario, we give a series of detection points, which are located above the containers, and there is no collision between two adjacent points" If there isn't collisions, it is never called the bias_RRT* algorithm, so why to compare to them. bias_RRT* is to obtain the detection points, which you already give.
Scenario 2: "the adjacent detection points are not all collision free, so the impact on the optimization and computational time is not as favorable as in scenario 1", how many detection points are not collision free?
You should change the intention and comparison of the paper. Because your proposal is interesting but you are performing an unfair comparison. It is also interesting when there are changes in the environment, hence, the previously known detection points are not longer valid.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsI believe the authors should test their algorithm against some basic heuristic approaches, to demonstrate the benefits of their solution. For example, a simple back-and-forth "lawnmower" trajectory could be used a comparison baseline.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors1) Consider defining acronyms before using them in the text. For instance, UAV has been used in the abstract without any definition.
2) Literature review is weak and needs to be improved. Consider discussing the following papers to improve the literature review:
[R1]. "Coverage Path Planning Methods Focusing on Energy Efficient and Cooperative Strategies for Unmanned Aerial Vehicles", 2022. [https://doi.org/10.3390/s22031235]
[R2]. "Constrained Control of UAVs in Geofencing Applications", 2018. [http://doi.org/10.1109/MED.2018.8443035]
[R3]. "Comparison of Parallel Genetic Algorithm and Particle Swarm Optimization for Real-Time UAV Path Planning", 2013. [http://doi.org/10.1109/TII.2012.2198665]
3) What is the difference(s) between the submitted manuscript and the authors' previous work "UAV Trajectory Planning in a Port Environment" 2020?
4) Figure 4 is too small and unclear. Consider replacing it with a legible one.
5) According to Figure 12, average deviation is around 20%. Is it acceptable for real-world implementation? Also, provide computing time of the proposed method and discuss why the proposed method is acceptable for real-time applications.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThe authors proposed a new path planning method for use in ports. However, the necessity and importance of this method has not been emphasized enough. I have some major suggestions:
1) “The bias_RRT* algorithm [14] has been used …”. Although the method name in study 14 was RRT-Connect, why was the expression “bias_RRT*” used?
2) The texts in Figure 1 must be in English. In addition, markings should be made in accordance with the explanation on this image.
3) “In order to optimize the application of the RTT*” there is a typo.
4) Figure 4 resolution should be increased.
5) Is the trajectory optimization suitable for real-time applications?
6) Is there any difference in the a and b images in Figure 9? The differences should be shown more clearly.
7) Path planning studies described in Section 2 should be expanded. Disadvantages of each method should be mentioned.
8) The advantages of the proposed method over previous path planning studies should be emphasized more clearly and broadly.
9) Algorithm should be added for the improved algorithm (Section 4.2).
10) Discussion section should be added.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsAuthors have addressed my comments satisfactorily.
Author Response
Thank you very much for your recognition and support of our paper revision. Your comments are very helpful to us in the process of our paper revision, which greatly improves the quality of our paper. Once again, we would like to express our sincere thanks and respect to you.
Reviewer 3 Report
Comments and Suggestions for AuthorsNo further comment.
Author Response
Thank you very much for your recognition and support of our paper revision. Your comments are very helpful to us in the process of our paper revision, which greatly improves the quality of our paper. Once again, we would like to express our sincere thanks and respect to you.
Reviewer 4 Report
Comments and Suggestions for AuthorsThe authors revised the article based on the comments. However, as the focus of this study is to develop a new RRT* variant, new and more RRT* studies should be added to section 2.1. For example, comparisons should be made with more recent and comprehensive studies such as F-RRT*, Quick-RRT*, GDRRT*.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 4 Report
Comments and Suggestions for AuthorsThe authors have improved the article. The article is acceptable.
