Submit to this Journal Review for this Journal Propose a Special Issue

Article Menu

Share Help Cite Discuss in SciProfiles

Open AccessArticle

Peer-Review Record

Optimizing Autonomous UAV Navigation with D* Algorithm for Sustainable Development

Sustainability 2024, 16(17), 7867; https://doi.org/10.3390/su16177867

by Pannee Suanpang^1,*

and Pitchaya Jamjuntr²

Reviewer 1: Anonymous

Reviewer 2: Anonymous

Reviewer 3:

Vadim Zhmud

Sustainability 2024, 16(17), 7867; https://doi.org/10.3390/su16177867

Submission received: 23 July 2024 / Revised: 28 August 2024 / Accepted: 2 September 2024 / Published: 9 September 2024

(This article belongs to the Topic Energy Management and Sustainable Development from Economic, Social and Environmental Aspects)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Review comments

Overall a good paper.

However, there are several weaknesses which need to be addressed in the revised submission.

#1. The introduction needs to be re-written

The authors need to clearly state what their objectives are in the first part of the paper. This has to be stated clearly

The second problem in the introduction section is that there is no clear statement of what the contribution is.

I would recommend the following style:

First state what the research problem is and in that context state what the research voids are in the literature that they are addressing.

Next they can state their own papers contribution.

#2. The literature review needs a major improvement .revised paper needs to be significantly more informative.

It is important for the authors to step back a bit and review papers related to UAV based monitoring approaches, any recent cyber physical conceptual frameworks or any systems implemented.

The current references have a number of papers from universities in china that are referenced to their web pages rather than journal or conference publications. I would recommend to also include papers in this topic from ACM, IEEE and other international conferences

Here are some recommended papers I would suggest:

Pl make sure to point the readers to review papers related to research in UAV path planning.

Here are some good survey papers:– I would recommend you also define what is the UAV pathplanning problem just like the way the authors have done in these survey papers.

Title: A Literature Review of UAV 3D Path Planning

https://scholar.google.com/scholar_url?url=https://www.researchgate.net/profile/Liang-Yang-52/publication/282744674_A_literature_review_of_UAV_3D_path_planning/links/57b6b32808aeddbf36e94f48/A-literature-review-of-UAV-3D-path-planning.pdf&hl=en&sa=X&ei=8TyYZqiQFpGP6rQPjKeIiAs&scisig=AFWwaeZ-E9657LHlKjogUXdJrBYD&oi=scholarr

Title: Path Planning for UAV Communication Networks: Related Technologies, Solutions, and Opportunities

https://dl.acm.org/doi/abs/10.1145/3560261

The authors need to recognize the importance of 3D simulation environments where UAV path plans can be studied virtually and validated or modified. Would recommend to search for such UAV papers dealing with simulation environments. They need to expand the literature review and relate the importance of their work which can be integrated with simulation frameworks as well as surveillance applications.

Here are some papers:

Autonomous UAV Surveillance in Complex Urban Environments

https://ieeexplore.ieee.org/document/5284858

A Conceptual framework for supporting UAV based cyber physical weather monitoring activities,

https://ieeexplore.ieee.org/document/8369588

UAV-enabled intelligent traffic policing and emergency response handling system for the smart city.

https://doi.org/10.1007/s00779-019-01297-y

regarding the later part of the paper; split your current conclusion into 2 sections:

first, have a section called Discussion of results. Then discuss them. Do not combine this with the conclusion.

pl make sure to expand the discussions of your research. Create sub-titles for the main points you want to discuss (one para for each point you want to discuss).

This discussion needs to be in detail and not just a summary.

In conclusion, pl address following:

summarize what the contributions of your research was.

discuss ideas for future research again in sub section or sub titles: currently these ideas are there but please expand them with sub titles for each new idea for research

Also explain how your approach dealing with 3D path planning approaches can become part of emerging digital twins or simulation based approaches for such path planning contexts.

Author Response

Dear Reviewer, the author extends sincere appreciation to the reviewers for their kindness, valuable suggestions, and constructive feedback, which have significantly enhanced the quality of our paper. In response to your recommendations, we have thoroughly revised and amended the manuscript. The changes are highlighted in blue, with additional information indicated in red within these highlights.

Overall a good paper.

However, there are several weaknesses which need to be addressed in the revised submission.

Reviewer: The introduction needs to be re-written

The authors need to clearly state what their objectives are in the first part of the paper. This has to be stated clearly. The second problem in the introduction section is that there is no clear statement of what the contribution is. I would recommend the following style:

First state what the research problem is and in that context state what the research voids are in the literature that they are addressing. Next they can state their own papers contribution.

Answer: Thank you for your valuable feedback. We appreciate your detailed suggestions for improving the introduction section of our paper.

Stating Objectives Clearly: We have revised the introduction to clearly articulate the objectives of our research at the beginning of the section. This ensures that the purpose of our study is immediately clear to the readers. (Lines 75-95)
Research Problem and Challenges: We have also restructured the introduction to clearly define the research problem and the specific gaps in the literature that our study aims to address. By identifying these research voids, we have provided a stronger context for the relevance of our work. (Lines 93-151)
Contribution: Following the identification of the research problem and gaps, we have explicitly stated the contributions of our paper. This highlights how our research fills these gaps and contributes to the broader academic discourse. (Lines 152-193)

We believe these changes enhance the clarity and impact of the introduction and better align it with the recommendations you provided. We hope that the revised introduction meets your expectations.

Reviewer: The literature review needs a major improvement revised paper needs to be significantly more informative. It is important for the authors to step back a bit and review papers related to UAV based monitoring approaches, any recent cyber physical conceptual frameworks or any systems implemented. The current references have a number of papers from universities in china that are referenced to their web pages rather than journal or conference publications. I would recommend to also include papers in this topic from ACM, IEEE and other international conferences. Here are some recommended papers I would suggest:

Pl make sure to point the readers to review papers related to research in UAV path planning.
Here are some good survey papers:– I would recommend you also define what is the UAV pathplanning problem just like the way the authors have done in these survey papers.
Title: A Literature Review of UAV 3D Path Planning

Title: Path Planning for UAV Communication Networks: Related Technologies, Solutions, and Opportunities

https://dl.acm.org/doi/abs/10.1145/3560261

The authors need to recognize the importance of 3D simulation environments where UAV path plans can be studied virtually and validated or modified. Would recommend to search for such UAV papers dealing with simulation environments. They need to expand the literature review and relate the importance of their work which can be integrated with simulation frameworks as well as surveillance applications.
Here are some papers: Autonomous UAV Surveillance in Complex Urban Environments

https://ieeexplore.ieee.org/document/5284858

A Conceptual framework for supporting UAV based cyber physical weather monitoring activities, https://ieeexplore.ieee.org/document/8369588
UAV-enabled intelligent traffic policing and emergency response handling system for the smart city. https://doi.org/1007/s00779-019-01297-y

Answers: Thank you for your detailed and insightful feedback regarding the literature review section of our paper. We recognize the importance of improving this section and have made substantial revisions based on your recommendations.

Broadening the Scope of the Literature Review: We have significantly expanded our literature review to include a more comprehensive examination of UAV-based monitoring approaches, recent developments in cyber-physical conceptual frameworks, and various systems that have been implemented. We have replaced references to non-peer-reviewed sources with peer-reviewed journal and conference publications, particularly from ACM, IEEE, and other reputable international conferences. We added your recommendation paper including (Lines 196-227)

A Literature Review of UAV 3D Path Planning [39]. (Lines 204-214)
Path Planning for UAV Communication Networks: Related Technologies, Solutions, and Opportunities [45]. (Lines 204-214)
Autonomous UAV Surveillance in Complex Urban Environments [46]. (Lines 204-214)
A Conceptual Framework for Supporting UAV-Based Cyber-Physical Weather Monitoring Activities [47]. (Lines 204-214)
UAV-enabled Intelligent Traffic Policing and Emergency Response Handling System for the Smart City [48]. (Lines 204-214)

We believe that these revisions have greatly strengthened the quality and informativeness of our literature review, addressing the concerns you raised. We are grateful for your guidance and are confident that these improvements will contribute to the overall impact of our paper.

Reviewer: Regarding the later part of the paper; split your current conclusion into 2 sections: first, have a section called Discussion of results. Then discuss them. Do not combine this with the conclusion. pl make sure to expand the discussions of your research. Create sub-titles for the main points you want to discuss (one para for each point you want to discuss). This discussion needs to be in detail and not just a summary.

Answers: Thank you for your valuable feedback. In response, we have revised the latter part of the paper by reorganizing the discussion into a more structured format with distinct sections: (Lines 1010-1102)

5.1 Interpretation of the Results: We have added a dedicated section titled "Discussion of Results," where the findings of our study are thoroughly analyzed and interpreted. This section provides a detailed examination of the implications and significance of the results, distinct from the conclusions.

5.2 Advantages and Disadvantages of the D Algorithm*: This section discusses the strengths and limitations of the D* Algorithm, offering a balanced evaluation of its performance in the context of our study.

5.3 Implications for Sustainable Development Applications: Here, we explore the broader implications of our research for sustainable development, highlighting how the findings can contribute to advancing sustainability practices.

5.4 Algorithmic Limitations: In this section, we address the specific limitations of the D* Algorithm, providing a critical assessment of its constraints and areas where improvements are needed.

5.5 Future Research Directions: Finally, we outline potential future research directions, identifying key areas where further investigation could enhance the effectiveness and applicability of the algorithm.

These changes aim to improve the clarity and depth of the discussion, ensuring that each aspect of the study is comprehensively addressed. We believe this restructuring strengthens the overall impact of the paper.

Reviewer: In conclusion, pl address following: summarize what the contributions of your research was. discuss ideas for future research again in sub section or sub titles: currently these ideas are there but please expand them with sub titles for each new idea for research.

Answers: Thank you for your constructive feedback. We have revised the conclusion section to better address your suggestions.

Summary of Contributions: We have added a concise summary of the key contributions of our research at the beginning of the conclusion. This section highlights the novel aspects and the impact of our study. (Lines 1104-1171)
Expanded Future Research Directions: To enhance clarity and depth, we have created distinct sub-sections within the conclusion, each dedicated to a specific idea for future research. These sub-sections allow for a more detailed discussion of potential areas for further investigation, providing a clearer roadmap for future work. (Lines 1186-1198)

We believe these changes will strengthen the conclusion by clearly articulating the contributions of our research and providing a structured approach to discussing future research directions. Thank you for your guidance.

Reviewer: Also explain how your approach dealing with 3D path planning approaches can become part of emerging digital twins or simulation based approaches for such path planning contexts

Answers: Thank you for your valuable feedback. We have incorporated information on 3D path planning in our future study section. We envision a future where the D* algorithm plays a pivotal role in autonomous UAV navigation, leading to safer and more efficient operations across diverse environments. By pursuing these research initiatives, we lay the groundwork for groundbreaking advancements that will shape the future of aerial operations. Finally, further research on the integration of digital twin technology with UAV 3D path planning should focus on leveraging real-time virtual replicas of physical systems to significantly enhance UAV navigation. By simulating and optimizing flight paths in complex environments, digital twins can substantially improve the adaptability, safety, and efficiency of UAV operations through continuous updates based on real-world data. This approach also addresses critical challenges, such as ensuring model accuracy and managing computational demands, ultimately leading to more reliable and effective UAV deployments [92]. (Lines 1195-1205)

The authors would like to extend their sincere gratitude to Reviewer for the invaluable suggestions to enhance the quality of this paper further. We also deeply appreciate the encouragement provided to the authors in presenting research on the application of advanced technologies. ?❤️?

Reviewer 2 Report

Comments and Suggestions for Authors

This article presents the integration of the D* algorithm into the navigation for autonomous UAVs. The D* algorithm was initially developed for ground robots. Integrating said algorithm into autonomous UAVs requires tackling the problems of path planning, aerial obstacles, and integration with UAV specific sensors and control systems. This article tackles these challenges. The experimental evaluation demonstrates the precision, adaptability, collision avoidance efficacy, and resilience of the approach.

Thank you for submitting your work to sustainability! I found this article very interesting and thought provoking, and I believe that it will be a good fit for the journal.

Positive points:

- The idea of integrating the D* algorithm into the field of unmanned UAVs is novel, with many interesting challenges

- A thorough evaluation of the approach by the authors demonstrates the efficacy and viability of the approach

- The article in general is well organized, and has sufficient background to bring a new reader up to speed

Author Response

Reviewer: This article presents the integration of the D* algorithm into the navigation for autonomous UAVs. The D* algorithm was initially developed for ground robots. Integrating said algorithm into autonomous UAVs requires tackling the problems of path planning, aerial obstacles, and integration with UAV specific sensors and control systems. This article tackles these challenges. The experimental evaluation demonstrates the precision, adaptability, collision avoidance efficacy, and resilience of the approach.

Reviewer: Thank you for submitting your work to sustainability! I found this article very interesting and thought provoking, and I believe that it will be a good fit for the journal.

Answer: Thank you very much, Reviewer, for thoroughly reading and reviewing our paper, as well as for the encouraging words you have shared with our team. We are deeply grateful for your appreciation and will take your compliments as motivation to continue improving our work to the best of our abilities.

Reviewer: Positive points:

- The idea of integrating the D* algorithm into the field of unmanned UAVs is novel, with many interesting challenges

Answer: Thank you for your insightful feedback. We appreciate your recognition of the novelty of integrating the D* algorithm into the field of unmanned UAVs. Indeed, the process presents many interesting challenges, which we aimed to address comprehensively in our study. Your comments motivate us to continue exploring and refining this innovative approach.

- A thorough evaluation of the approach by the authors demonstrates the efficacy and viability of the approach

Answer: Thank you for your thoughtful review. We are pleased that the thorough evaluation of our approach effectively demonstrates its efficacy and viability. Your recognition of our efforts in this regard is greatly appreciated and encourages us to continue advancing our research.

- The article in general is well organized, and has sufficient background to bring a new reader up to speed

Answer: Thank you for your positive feedback. We're glad to hear that you found the article well-organized and that the background information provided was sufficient to bring new readers up to speed. Your comments reinforce our commitment to clarity and accessibility in presenting our research.

Reviewer 3 Report

Comments and Suggestions for Authors

Autonomous navigation for unmanned aerial vehicles (UAVs) has become a critical tool in a variety of industries, from agriculture, delivery services, surveillance to search and rescue. However, navigating UAVs in dynamic and unknown environments remains a challenge. This paper examines the application of the D* algorithm, a well-known path planning method based on robotics and artificial intelligence, and compares it with other algorithms such as A* and RRT* to enhance autonomous navigation capabilities in the context of UAVs for sustainable development. The main problem addressed in this paper revolves around improving the navigation efficiency, safety and adaptability of UAVs in dynamic environments. The research methodology includes the integration of the D* algorithm into the UAV navigation system, which allows real-time adjustment and planning of the path taking into account dynamic obstacles and changing terrain conditions. The experimentation phase takes place in simulated environments designed to simulate real-life scenarios and problems. Extensive data collection, rigorous analysis, and performance evaluations paint a clear picture of the performance of the D* algorithm compared to other navigation methods such as A* and RRT*. The main findings indicate that the D* algorithm offers a compelling solution by providing UAVs with efficient, safe and adaptive navigation capabilities. The results demonstrate a 92% improvement in path planning efficiency, a 5% reduction in collision rates and an increase in safety margins by 2.3 meters. The contribution of this paper is to demonstrate the practical effectiveness of the D* algorithm, along with comparisons with A* and RRT*, in improving the autonomous navigation of UAVs, promoting the development of aerial systems. In particular, this study provides insight into the strengths and limitations of each algorithm, offering valuable guidance for researchers and practitioners to select the most appropriate path planning approach for their UAV applications. The implications of this research extend far and wide, with potential applications in industries such as agriculture, surveillance, disaster response and more for sustainability.

Instead of the phrase: “ method rooted in robotics and artificial intelligence " the phrase suggested is: " method" rooted on artificial intelligence and widely used in robotics ”, since it should be distinguished: this method is not based on robotics, but is used in it, while the method is based on algorithms that allow it to be classified as artificial intelligence, although, strictly speaking, there is not so much intelligence in the method, it is based on a completely simple and unambiguous algorithm.

IN lines 66-67 are given proposal that more corresponds genre entertaining or popular science literature , but V scientific articles such literary rpm Not welcome : “In the pages that follow, we embark on a journey through the integration of the D* algorithm into UAV navigation systems.” Although the reviewer personally, as a person who is not alien to literature, likes this proposal, it is still recommended to adhere to the standard language of scientific articles.

All three paragraphs in the section “Problem Statement” do not refer to problem statement . The first two paragraphs should still be a description of the situation, and the last paragraph should be a summary of the achievements described in the article.

The statement of the problem should be given in terms: There is a certain situation, there are certain data, it is required to provide some not yet existing actions or a new status of the situation, after which there is a section “methods for solving the problem.”

Here we continue the discussion in terms of: “there are certain shortcomings” and “our article solves some problems.” A description of what was obtained is appropriate in the abstract and in the conclusion, but not in the “Problem Statement” section.

Statement V lines 79 – 80: “PID controllers, have demonstrated limitations in providing optimal paths ” justified . PID controllers are not designed to control the optimal path; they are designed to return the vehicle to a given trajectory if it deviates from this trajectory due to the influence of various interfering factors. A system with PID controllers is responsible only for the accuracy of following the selected trajectory, and not for the choice of this trajectory. PID controllers may have drawbacks in some cases, however, it has been proven by repeated practice that these are the most effective and most easily customizable controllers from the entire set of control structures and algorithms, of which there are already more than three dozen, but trajectory selection algorithms should not be counted among them, and these devices cannot be classified as such algorithms; these are different levels of hierarchy in the control of an autonomous vehicle. Blaming PID controllers for shortcomings in trajectory formation is the same as blaming an artist’s brush for shortcomings in choosing the subject of a painting. At the same time, it is completely possible that the D* algorithm is effective as one of the most effective algorithms for adaptive construction and restructuring of the required trajectory, and even, perhaps, it is the best algorithm, and it is very possible that its use reduces the load on PID controllers due to restructuring the trajectory to a smoother one.

Section 1.3 “Purpose of the study” comes after section 1.1. "Formulation of the problem". This is unusual. Usually, a goal is a more general concept that describes the problem as a whole; the goal is achieved by solving many problems, and not all of these problems are solved by the article. For example, in order to provide education, a student is taught mathematics. Mathematics is only part of education. Or, in order to better equip the workshop, additional tools are purchased. And the tasks are very specific tasks arising from the goal, which are exactly solved in the article. The set tasks must be solved in the article, but the goal is not achieved as a result of only what is described in the article, its solution is only approached. Therefore, it is not the goal that is a consequence of the tasks, but the tasks are the consequence of the set goal, and after formulating the tasks, one should not talk about the goal.

At the same time, demonstrating the advantages of some algorithm is even lower than the research objectives. The task is, for example, the calculation of specific trajectories, and the operation of the algorithm can be demonstrated on the simplest trajectory, even on a fictitious trajectory, very far from real problems. So it should not be said that the purpose of the article is to demonstrate the effectiveness of the algorithm. If an algorithm already exists, then its authors have already demonstrated some of its advantages. At a minimum, the goal of the article may be to compare two or more algorithms using an example of a practically useful problem. Moreover, there cannot be two or more goals. The goal is always the same, there are many tasks. Section 3.1 “Purpose of the Study” talks about two different goals, these are not goals, but objectives.

IN Lines 175 – 176 say : “This paper contributes a nuanced understanding of the D* algorithm's potential to 175 enhance autonomous UAV navigation within the simulation paradigm.” A scientific article is not a textbook. Understanding of some scientific facts may occur in one group of researchers, but not in another group of research. A scientific article is not intended to improve someone's understanding of a problem or methods for solving it, but to propose new, improve, or develop existing methods and demonstrate (that is, prove) the usefulness of these solutions. Instead of “improving understanding”, it is better to use terms such as: “confirm”, “demonstrate”, “prove”, “research” and so on.

Figure 2 does not show the “Algorithm”, but the result of this algorithm when solving a two-dimensional problem. Algorithms are shown differently - in the form of a block diagram, as in Figure 4.

Figure 7 should be redone, or it is doubtful that it can be retained in the article. The authors showed the use of three different algorithms to solve three different problems. Or it was necessary to show the solution of the same problem using three different algorithms, which would be very interesting, visual and useful, and at the same time it would be possible to visually compare the difference in the lengths of trajectories and show the difference in the number of operations in the solution. If the authors do not have the potential to solve the same problem, it would be necessary to admit that the authors did not study all three algorithms in detail, but worked only with one of them, and then illustrations of the work of the other two algorithms are not appropriate. Moreover, the first of the three figures in the general Figure 7 exactly repeats Figure 4, which, firstly, is unnecessary, and secondly, raises suspicions that the authors were unable to solve some other problem using this algorithm.

If, nevertheless, for each individual task only one of the three algorithms is more effective, then it is not legitimate to say that the algorithm under discussion is better than others, but it should be stated that for each task it is necessary to choose a different algorithm, which, apparently, is not the case . In addition, there is a clear dissonance between the fact that in the introduction the authors said that controlling unmanned aerial vehicles is a three-dimensional problem, and the illustrations show solutions to a two-dimensional problem. The formulation of the problem should always correspond to the results obtained.

The same applies to Figures 8, 9, 10. In all these figures, only tasks within interconnected pairs are modified - for each algorithm there is a separate task, which is slightly modified and complicated, but there is no commonality. Then it turns out unconvincing - just like comparing a hammer, a screwdriver and a medical syringe, claiming that at the same time the effectiveness of these three tools is being compared. Either the tools should be compared when solving identical problems, or it should be recognized that there is simply no comparison.

Due to the noted shortcomings, the conclusion to the article should also be properly edited.

In general, the authors conducted an interesting study; after improving the formulation of the problem and the way the results are described and compared with each other, the article requires additional review; if the improvements are sufficient, the article can be published.

Perhaps a bibliography of 90 sources is excessive? Did the authors actually study all of these 90 sources and use the information found in them to write their article?

Author Response

Autonomous navigation for unmanned aerial vehicles (UAVs) has become a critical tool in a variety of industries, from agriculture, delivery services, surveillance to search and rescue. However, navigating UAVs in dynamic and unknown environments remains a challenge. This paper examines the application of the D* algorithm, a well-known path planning method based on robotics and artificial intelligence and compares it with other algorithms such as A* and RRT* to enhance autonomous navigation capabilities in the context of UAVs for sustainable development. The main problem addressed in this paper revolves around improving the navigation efficiency, safety and adaptability of UAVs in dynamic environments. The research methodology includes the integration of the D* algorithm into the UAV navigation system, which allows real-time adjustment and planning of the path taking into account dynamic obstacles and changing terrain conditions. The experimentation phase takes place in simulated environments designed to simulate real-life scenarios and problems. Extensive data collection, rigorous analysis, and performance evaluations paint a clear picture of the performance of the D* algorithm compared to other navigation methods such as A* and RRT*. The main findings indicate that the D* algorithm offers a compelling solution by providing UAVs with efficient, safe and adaptive navigation capabilities. The results demonstrate a 92% improvement in path planning efficiency, a 5% reduction in collision rates and an increase in safety margins by 2.3 meters. The contribution of this paper is to demonstrate the practical effectiveness of the D* algorithm, along with comparisons with A* and RRT*, in improving the autonomous navigation of UAVs, promoting the development of aerial systems. In particular, this study provides insight into the strengths and limitations of each algorithm, offering valuable guidance for researchers and practitioners to select the most appropriate path planning approach for their UAV applications. The implications of this research extend far and wide, with potential applications in industries such as agriculture, surveillance, disaster response and more for sustainability.

Reviewer: Instead of the phrase: “ method rooted in robotics and artificial intelligence " the phrase suggested is: " method" rooted on artificial intelligence and widely used in robotics ”, since it should be distinguished: this method is not based on robotics, but is used in it, while the method is based on algorithms that allow it to be classified as artificial intelligence, although, strictly speaking, there is not so much intelligence in the method, it is based on a completely simple and unambiguous algorithm.

Answer: Thank you for the suggestion. The proposed phrase, "method rooted in artificial intelligence and widely used in robotics," is indeed more precise. It correctly clarifies that the method is not based on robotics but is applied within the field, with its foundation in algorithms that classify it as artificial intelligence. Although the method relies on a straightforward and unambiguous algorithm, the classification as AI is appropriate. We appreciate this refinement to enhance the accuracy of our description. (Lines 12-13)

Reviewer: IN lines 66-67 are given proposal that more corresponds genre entertaining or popular science literature , but V scientific articles such literary rpm Not welcome : “In the pages that follow, we embark on a journey through the integration of the D* algorithm into UAV navigation systems.” Although the reviewer personally, as a person who is not alien to literature, likes this proposal, it is still recommended to adhere to the standard language of scientific articles.

Answer: Thank you for your valuable feedback. We appreciate your suggestion regarding the use of more formal and standard language in scientific articles. We will revise the phrasing to better align with the conventions of academic writing, ensuring clarity and precision in our presentation. Your guidance is greatly appreciated. We have revised the statement to: “In the following sections, we explore the integration of the D* algorithm into UAV navigation systems." (Line 68-69)

Reviewer: All three paragraphs in the section “Problem Statement” do not refer to problem statement . The first two paragraphs should still be a description of the situation, and the last paragraph should be a summary of the achievements described in the article.

Answer: Thank you for pointing out this issue. We recognize that the paragraphs in the “Problem Statement” section may not directly address the problem as intended. We will revise the first two paragraphs to ensure they accurately describe the situation, and we will adjust the last paragraph to serve as a summary of the achievements discussed in the article. Your feedback is invaluable in helping us improve the clarity and structure of our work. (Line 76-101)

Reviewer: The statement of the problem should be given in terms: There is a certain situation, there are certain data, it is required to provide some not yet existing actions or a new status of the situation, after which there is a section “methods for solving the problem.”

Answer: Thank you for your guidance. The revised problem statement now adheres to the recommended structure:

“There is a situation where UAVs are increasingly being integrated into various industries, but current navigation systems, such as waypoint following and PID controllers, struggle to operate effectively in dynamic and unpredictable environments. The available data shows significant limitations in these systems, particularly in dealing with complex terrains and obstacles. Therefore, there is a need to develop new autonomous navigation actions that do not yet exist, specifically a framework that enhances UAVs' adaptability, allows real-time adjustments, and ensures collision-free path planning. Following this, the paper will discuss the methods for solving this problem by exploring the potential of the D* algorithm as a solution to these challenges.” (Lines 92-100)

Reviewer: Here we continue the discussion in terms of: “there are certain shortcomings” and “our article solves some problems.” A description of what was obtained is appropriate in the abstract and in the conclusion, but not in the “Problem Statement” section.

Answer: Thank you for your clarification. We understand that the "Problem Statement" section should concentrate on outlining the shortcomings and specific challenges our article addresses, rather than presenting the outcomes. We have already adjusted the problem statement to reflect this focus and have reserved the detailed phrases " This article addresses certain challenges and contributes by demonstrating the practical effectiveness of the D* algorithm, alongside comparisons with A* and RRT*, in enhancing autonomous UAV navigation and advancing aerial systems.” in abstract (Lines 25-27). Also, we change the conclusion section by provide this “While our article solves some problems through this contribution by offers valuable insights into the potential of the D* algorithm in advancing autonomous UAV navigation, it is essential to acknowledge and address certain limitations that lay the groundwork for future research endeavors” (Line 1054-1057)

Reviewer: Statement V lines 79 – 80: “PID controllers, have demonstrated limitations in providing optimal paths ” justified . PID controllers are not designed to control the optimal path; they are designed to return the vehicle to a given trajectory if it deviates from this trajectory due to the influence of various interfering factors. A system with PID controllers is responsible only for the accuracy of following the selected trajectory, and not for the choice of this trajectory. PID controllers may have drawbacks in some cases, however, it has been proven by repeated practice that these are the most effective and most easily customizable controllers from the entire set of control structures and algorithms, of which there are already more than three dozen, but trajectory selection algorithms should not be counted among them, and these devices cannot be classified as such algorithms; these are different levels of hierarchy in the control of an autonomous vehicle. Blaming PID controllers for shortcomings in trajectory formation is the same as blaming an artist’s brush for shortcomings in choosing the subject of a painting. At the same time, it is completely possible that the D* algorithm is effective as one of the most effective algorithms for adaptive construction and restructuring of the required trajectory, and even, perhaps, it is the best algorithm, and it is very possible that its use reduces the load on PID controllers due to restructuring the trajectory to a smoother one.

Answer: Thank you for your detailed explanation. You are correct that PID controllers are not intended to optimize path selection but rather to maintain accuracy in following a predefined trajectory. They are not responsible for choosing the path itself, and their effectiveness as controllers is well-established due to their ease of customization and reliability. We acknowledge that it is not appropriate to criticize PID controllers for limitations in trajectory formation, as this function is outside their intended scope. Instead, the D* algorithm, which excels in adaptive path planning, may complement PID controllers by creating smoother trajectories, thereby reducing the controllers' workload. We appreciate this perspective and will consider it in the context of our research. (Lines 78-83)

Reviewer: Section 1.3 “Purpose of the study” comes after section 1.1. "Formulation of the problem". This is unusual. Usually, a goal is a more general concept that describes the problem as a whole; the goal is achieved by solving many problems, and not all of these problems are solved by the article. For example, in order to provide education, a student is taught mathematics. Mathematics is only part of education. Or, in order to better equip the workshop, additional tools are purchased. And the tasks are very specific tasks arising from the goal, which are exactly solved in the article. The set tasks must be solved in the article, but the goal is not achieved as a result of only what is described in the article, its solution is only approached. Therefore, it is not the goal that is a consequence of the tasks, but the tasks are the consequence of the set goal, and after formulating the tasks, one should not talk about the goal.

Answer: Thank you for your feedback. We have updated the section title from "Objective of the Study" to "Purpose of the Study," which is a more appropriate term in the context of an academic paper. This section will now align with the typical structure of the Introduction, which generally includes:

Statement of the Problem: Clearly defines the problem or gap in knowledge that the study aims to address. (Lines 75-100)
Challenges and Solution Proposal: Identification of key challenges and proposed solutions. (Lines 102-133)
Purpose of the Study: Outlines the overall goal of the research and what it seeks to achieve. (Lines 135-154)
Significance of the Study: Explains the importance of research and its potential contributions to the field. (Lines 155-196)

This adjustment ensures that our paper adheres to the standard format, improving clarity and alignment with academic conventions.

Reviewer: At the same time, demonstrating the advantages of some algorithm is even lower than the research objectives. The task is, for example, the calculation of specific trajectories, and the operation of the algorithm can be demonstrated on the simplest trajectory, even on a fictitious trajectory, very far from real problems. So it should not be said that the purpose of the article is to demonstrate the effectiveness of the algorithm. If an algorithm already exists, then its authors have already demonstrated some of its advantages. At a minimum, the goal of the article may be to compare two or more algorithms using an example of a practically useful problem. Moreover, there cannot be two or more goals. The goal is always the same, there are many tasks. Section 3.1 “Purpose of the Study” talks about two different goals, these are not goals, but objectives.

Answer: We have revised the objectives to align with the content and experimental results presented in the article. The objective is now stated as follows: "This paper aims to explore the application of the D* algorithm, a prominent path-planning method rooted in artificial intelligence and widely used in robotics, alongside comparisons with other algorithms, such as A* and RRT*, to augment autonomous navigation capabilities in UAVs with implications for sustainable development." (Lines 136-143)

Additionally, we have included content to enhance the reader's understanding: "Moreover, this research aims to provide practical insights into the integration process of the D* algorithm into UAV navigation systems. This involves meticulously detailing the algorithmic considerations and hardware configurations necessary for seamless implementation. By addressing the technical aspects, our aim is to facilitate the adoption of the D* algorithm by UAV developers, researchers, and practitioners. This secondary objective complements the primary goal by ensuring that the transformative potential of the D* algorithm is not only understood but also practically applied, contributing to the advancement of autonomous UAV capabilities in real-world scenarios [33][34]." (Lines 144-151)

Reviewer: IN Lines 175 – 176 say : “This paper contributes a nuanced understanding of the D* algorithm's potential to 175 enhance autonomous UAV navigation within the simulation paradigm.” A scientific article is not a textbook. Understanding of some scientific facts may occur in one group of researchers, but not in another group of research. A scientific article is not intended to improve someone's understanding of a problem or methods for solving it, but to propose new, improve, or develop existing methods and demonstrate (that is, prove) the usefulness of these solutions. Instead of “improving understanding”, it is better to use terms such as: “confirm”, “demonstrate”, “prove”, “research” and so on.

Answer: Thank you for your feedback. We acknowledge that a scientific article's primary purpose is to propose, improve, or develop methods and to demonstrate their effectiveness, rather than simply improving understanding. We have revised the language in our paper accordingly, replacing "improving understanding" with more appropriate terms such as "demonstrate" to better align with the objectives of scientific research. (Lies 176-178)

Reviewer: Figure 2 does not show the “Algorithm”, but the result of this algorithm when solving a two-dimensional problem. Algorithms are shown differently - in the form of a block diagram, as in Figure 4.

Answer: Thank you for your observation. You are correct that Figure 2 does not depict the “Algorithm” itself but rather the result of the algorithm when applied to a two-dimensional problem. Typically, algorithms are presented in the form of a block diagram, as shown in Figure 4, which visually represents the logical flow and steps of the algorithm. Figure 2, on the other hand, illustrates the output or solution generated by the algorithm in a specific scenario, such as the pathfinding process in a grid with obstacles. (Lines 281-286)

Reviewer: Figure 7 should be redone, or it is doubtful that it can be retained in the article. The authors showed the use of three different algorithms to solve three different problems. Or it was necessary to show the solution of the same problem using three different algorithms, which would be very interesting, visual and useful, and at the same time it would be possible to visually compare the difference in the lengths of trajectories and show the difference in the number of operations in the solution. If the authors do not have the potential to solve the same problem, it would be necessary to admit that the authors did not study all three algorithms in detail, but worked only with one of them, and then illustrations of the work of the other two algorithms are not appropriate. Moreover, the first of the three figures in the general Figure 7 exactly repeats Figure 4, which, firstly, is unnecessary, and secondly, raises suspicions that the authors were unable to solve some other problem using this algorithm.

Answer: Thank you for your valuable feedback. We have carefully considered your comments and have conducted additional simulations to address the concerns raised. We have redone Figure 7 to reflect the results of applying the three different algorithms to the same problem. This revision allows for a direct visual comparison of the algorithms, including differences in the lengths of trajectories and the number of operations required for each solution.

We believe that this updated figure aligns with your suggestion and enhances the clarity and usefulness of the visual comparison. Additionally, we have removed the redundant repetition of Figure 4 within Figure 7 to avoid any confusion and to ensure that the illustrations are both relevant and accurate. (Lines 868-868)

We appreciate your guidance in improving the quality of our manuscript and are confident that these revisions will contribute to a more robust presentation of our research findings.

Reviewer: If, nevertheless, for each individual task only one of the three algorithms is more effective, then it is not legitimate to say that the algorithm under discussion is better than others, but it should be stated that for each task it is necessary to choose a different algorithm, which, apparently, is not the case . In addition, there is a clear dissonance between the fact that in the introduction the authors said that controlling unmanned aerial vehicles is a three-dimensional problem, and the illustrations show solutions to a two-dimensional problem. The formulation of the problem should always correspond to the results obtained.

Answer: Thank you for your insightful feedback. We greatly appreciate the opportunity to address the issues you have highlighted. To improve the clarity and accuracy of our manuscript, we will revise the performance comparison to emphasize that the selection of algorithms should be tailored to the specific task and scenario at hand. (Lines 825-853)

Effectiveness of Algorithms for Specific Tasks: Our experimental results indicate that while the D* algorithm generally performs well across various scenarios, there are cases where other algorithms, such as A* or RRT*, may deliver more competitive or superior performance depending on the task. For instance, in highly cluttered environments, the A* algorithm may produce more optimal paths in terms of computational efficiency, whereas the RRT* algorithm excels in dynamically changing spaces. This variability highlights that no single algorithm is universally superior for all tasks. Therefore, it is essential to select the most appropriate algorithm based on the specific context and requirements of each task.
Dimensionality of the Problem: We acknowledge the discrepancy between the problem formulation presented in the introduction, which describes controlling unmanned aerial vehicles (UAVs) as a three-dimensional challenge, and the two-dimensional scenarios used in our initial experiments. The two-dimensional approach was intended as a preliminary step to simplify the problem and test fundamental aspects of the algorithms in a controlled environment. To address this, we will update the manuscript to clarify the scope of these initial experiments and outline our plan to extend the research to three-dimensional scenarios. This approach will ensure that the problem formulation aligns with both the results obtained and our future research directions.
In summary, we will provide a more nuanced analysis of each algorithm's relative strengths, highlighting the contexts where each may be most effective. The current focus on two-dimensional scenarios will be clearly acknowledged, and our roadmap for extending the research to three-dimensional problems will be articulated. This will ensure consistency between the problem formulation, the experimental results, and the future direction of the research.

Reviewer: The same applies to Figures 8, 9, 10. In all these figures, only tasks within interconnected pairs are modified - for each algorithm there is a separate task, which is slightly modified and complicated, but there is no commonality. Then it turns out unconvincing - just like comparing a hammer, a screwdriver and a medical syringe, claiming that at the same time the effectiveness of these three tools is being compared. Either the tools should be compared when solving identical problems, or it should be recognized that there is simply no comparison.

Answer: Thank you for your insightful feedback. In response, we have revised Figures 8-10 and conducted additional simulations based on your recommendations. The updated figures now present a direct comparison of the performance of each algorithm on identical tasks, allowing for a more meaningful and precise evaluation. We are confident that these revisions will greatly enhance the clarity and significance of our findings. (Lines 865-974, 884-895, 901-910, 919-927,940-949)

Reviewer: Due to the noted shortcomings, the conclusion to the article should also be properly edited.

Answer: Thank you for your constructive feedback. We acknowledge the need to address the shortcomings you have identified. In response, we will carefully revise the conclusion of the article to better reflect the limitations and to emphasize the key findings while also outlining the future research directions that can address these challenges. Our goal is to ensure that the conclusion provides a balanced and comprehensive summary of our work, acknowledging both the contributions and areas for improvement. We appreciate your guidance in helping us improve the quality of our manuscript. (Lines 1086-1097,1154-1180)

Reviewer: In general, the authors conducted an interesting study; after improving the formulation of the problem and the way the results are described and compared with each other, the article requires additional review; if the improvements are sufficient, the article can be published.

Answer: Thank you for your positive assessment of our study and for your constructive suggestions. We appreciate your feedback and will make the necessary improvements, particularly in refining the problem formulation and enhancing the way the results are described and compared. We are committed to addressing these areas to meet the standards required for publication. Once the revisions are complete, we welcome the opportunity for an additional review to ensure that the article is of the highest quality. We are confident that with these enhancements, the article will be ready for publication.

Reviewer: Perhaps a bibliography of 90 sources is excessive? Did the authors actually study all of these 90 sources and use the information found in them to write their article?

Answer: Thank you for your observation. We understand your concern regarding the number of sources cited in our bibliography. Each of the 90 sources was carefully reviewed and selected for their relevance and contribution to the insights and background information that informed our research and analysis. However, we have already rechecked and revisited the bibliography to ensure that only the most pertinent and directly referenced sources are included, streamlining the list to better reflect the core references that contributed to the development of our article. We appreciate your feedback and will make the necessary adjustments accordingly.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

As previously ordered, the PID controller is not a navigation method, it is a method of ensuring efficient movement along a selected trajectory. It is necessary to clearly distinguish the method of forming a desired or correct trajectory from the method of its implementation. In the considered phrase in lines 94–100, the understanding of this fact is not taken into account. The authors ignored the wonderful reviewers.

A comparison of the three methods, performed in the form of a histogram of the path length for different problems, shows that the method under discussion has an advantage, but it is always very small compared to the path length as a whole. If this algorithm has some disadvantages, then its stated advantage is not enough to consider it the best method. If there are no disadvantages, then this method is the best, but this statement is not reliable enough. Surprisingly, the differences are always approximately the same in magnitude, even in the case when the value itself differs by a multiple (the path length differs by two times, and the advantages of the method differ by the same absolute difference in the path length). This is strange, because if, for example, the path increases by two times, then the savings in path length due to a more successful algorithm, as expected, should differ by two times. Apparently, the statement about “significant advantages” in lines 1013-1014 is somewhat exaggerated, overly optimistic.

Earlier, the reviewer already pointed out that it is the most indicative. There would be a comparison of three aglorhythms, even if the regeneration of one and the same task was compared with three different algorithms. In this case, some corrections have been made in the article, but they are sufficient, because there is still a doubt that the same tasks were solved.

Let's turn to figure 8. The name has several rows and several columns of illustration. If each row corresponds to a different method, then, apparently, it should be expected that the solution to the same problem is shown in the boxes of each column. No, that's what it is. If you compare the illustrations, placed one under the other, you can see that the tasks, placed one under the other, are completely different. In the bottom line of the drawings, there are many obstacles that are located far from the optimal path and do not affect the movement. There is no such thing in other lines. In the two upper lines, the movement is from left to top to right to bottom. In the bottom row, the movement is from left to lower, up to right. Tasks clearly different. That's why the author of the article did not take into account the reviewer's comments. Thus, the illustrations are either not related to the comparison table, or they show only part of the tasks, and not the whole task, or different tasks were solved by different methods, and therefore all this is not visible, inappropriate.

Author Response

Dear Reviewer, the author extends sincere appreciation to the reviewers for their kindness, valuable suggestions, and constructive feedback, which have significantly enhanced the quality of our paper. In response to your recommendations, we have thoroughly revised and amended the manuscript. The changes are highlighted in yellow, with additional information indicated in yellow within these highlights.

Reviewer: As previously ordered, the PID controller is not a navigation method, it is a method of ensuring efficient movement along a selected trajectory. It is necessary to clearly distinguish the method of forming a desired or correct trajectory from the method of its implementation. In the considered phrase in lines 94–100, the understanding of this fact is not taken into account. The authors ignored the wonderful reviewers.

Answer: Thank you very much for your insightful comments regarding the PID controller. We have revised the manuscript according to your suggestions. Specifically, we have clarified the distinction between navigation methods and control methods. In the revised text, we have highlighted that the rapid expansion of UAV applications across various industries has underscored the critical need for highly adaptable and reliable autonomous navigation systems. Current navigation methods, such as waypoint following, often fall short in dynamic and unpredictable environments, particularly when dealing with complex terrains and obstacles. While control methods like PID controllers are crucial for ensuring efficient movement along a selected trajectory, they do not address the core issue of determining that trajectory, especially in real-time and in the presence of unforeseen obstacles. Therefore, the primary challenge lies in how UAVs can achieve autonomous navigation that is both highly adaptable and capable of making real-time adjustments to ensure collision-free path planning in these challenging and unknown environments. (Lines 94-121)

We hope that these revisions adequately address your concerns, and we sincerely appreciate your valuable feedback.

Reviewer: A comparison of the three methods, performed in the form of a histogram of the path length for different problems, shows that the method under discussion has an advantage, but it is always very small compared to the path length as a whole. If this algorithm has some disadvantages, then its stated advantage is not enough to consider it the best method. If there are no disadvantages, then this method is the best, but this statement is not reliable enough. Surprisingly, the differences are always approximately the same in magnitude, even in the case when the value itself differs by a multiple (the path length differs by two times, and the advantages of the method differ by the same absolute difference in the path length). This is strange, because if, for example, the path increases by two times, then the savings in path length due to a more successful algorithm, as expected, should differ by two times. Apparently, the statement about “significant advantages” in lines 1013-1014 is somewhat exaggerated, overly optimistic.

Answer: Thank you very much for your suggestions regarding the comparison of the three methods. We have provided a detailed explanation regarding the comparison. We understand the concerns raised about the scale of the advantages presented by the method under discussion. While our analysis showed some consistent benefits, we acknowledge that the magnitude of these advantages may appear relatively small when compared to the overall path length. We also recognize that the proportionality of the savings in path length to the overall path length, as pointed out, does not align with the expected outcome, which raises valid questions about the interpretation of the results. In light of this, we have revisited the analysis and revised our statements to ensure they more accurately reflect the observed data. Specifically, we have tempered the language used in lines 1039-1046 to better represent the practical significance of the advantages identified, acknowledging that while there are benefits, they may not be as significant as initially suggested. Furthermore, we have added additional context to explain the unexpected consistency in the magnitude of differences, providing a more balanced view of the algorithm's performance. (Lines 1029-1041)

We appreciate your valuable insights, which have helped us refine our discussion and presentation of the results.

Reviewer: Earlier, the reviewer already pointed out that it is the most indicative. There would be a comparison of three algorithms, even if the regeneration of one and the same task was compared with three different algorithms. In this case, some corrections have been made in the article, but they are sufficient, because there is still a doubt that the same tasks were solved.

Answer: Thank you very much for your suggestions regarding the comparison of the algorithms. We would like to clarify that the comparison of the D*, A*, and RRT* algorithms was indeed conducted on the same set of tasks. Specifically, all three algorithms were tested under identical conditions, including the same environment, obstacle configurations, dynamic changes, and other relevant parameters. This was done to ensure a fair and accurate comparison of their performance. To address the concern and ensure clarity, we have further detailed the experimental setup and methodology in the revised manuscript. Specifically, we have explicitly stated that each task scenario was consistently regenerated across all three algorithms, maintaining identical conditions and environments throughout the experiments. This approach guarantees that the comparison of results is both valid and reliable, with no variations influencing the outcomes. (Lines 910-924).

We hope this clarification resolves any remaining doubts regarding the comparability of the algorithms in our study.

Reviewer: Let's turn to figure 8. The name has several rows and several columns of illustration. If each row corresponds to a different method, then, apparently, it should be expected that the solution to the same problem is shown in the boxes of each column. No, that's what it is. If you compare the illustrations, placed one under the other, you can see that the tasks, placed one under the other, are completely different. In the bottom line of the drawings, there are many obstacles that are located far from the optimal path and do not affect the movement. There is no such thing in other lines. In the two upper lines, the movement is from left to top to right to bottom. In the bottom row, the movement is from left to lower, up to right. Tasks clearly different. That's why the author of the article did not take into account the reviewer's comments. Thus, the illustrations are either not related to the comparison table, or they show only part of the tasks, and not the whole task, or different tasks were solved by different methods, and therefore all this is not visible, inappropriate.

Answer: Thank you once again for your insightful suggestions regarding the improvement of Figure 8. (Lines 910-913) We have carefully revised the figure to more effectively illustrate the comparison of the D*, A*, and RRT* algorithms, all of which were evaluated under identical dynamic obstacle conditions. The updated figure now clearly demonstrates the superior performance of the D* algorithm, particularly in terms of path efficiency and collision rate. The graph underscores the adaptability of D* in dynamically changing environments, where it maintains a lower collision rate and a larger safety margin compared to A* and RRT*, positioning it as a reliable choice for real-time obstacle interaction. We trust that this revised figure more accurately reflects the comparison and fully addresses the concerns you raised. Thank you again for your valuable feedback.

The authors wish to express their sincere gratitude to the Reviewer for the invaluable suggestions, which have greatly contributed to the further enhancement of this paper's quality. We also deeply appreciate the encouragement in advancing our research on the application of advanced technologies. ?❤️?

Round 3

Reviewer 3 Report

Comments and Suggestions for Authors

The edited variant of the paper is OK.

Article Menu

Optimizing Autonomous UAV Navigation with D* Algorithm for Sustainable Development

Further Information

Guidelines

MDPI Initiatives

Follow MDPI