An Efficient Autonomous Exploration Framework for Unmanned Surface Vehicles in Unknown Waters

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Abstract:

·         Briefly define USV and state the motivation/challenges.

·         Concisely summarize the proposed solution (strategy and framework).

·         Highlight key contributions and improvements over existing methods.

·         Mention experiments and main results.

·         Quantify key results (e.g. "achieved an X% efficiency improvement") to concisely communicate findings.

Introduction:

·         Provide more context on USV exploration applications and challenges.

·         Summarize state-of-the-art methods and their limitations.

·         Clearly define the research problem/objectives.

·         Outline the proposed solution and contributions.

·         Preview the organization of the paper.

·         Cite relevant surveys/review papers to help position the work within the broader field.

Related Work:

·         Categorize existing methods as frontier-based, sampling-based, and hybrid.

·         Critically analyze each approach and discuss limitations.

·         Highlight how the proposed solution advances the literature.

·         When discussing methods, indicate the year of publication to establish chronology. Cluster by topic (e.g. frontier vs sampling) before critiquing approaches.

Problem Formulation:

·         Define environment parameters, sensor model, motion constraints.

·         Formally state the exploration and path planning problems.

·         Define variables and parameters scientifically using notation/terminology from the field.

Methodology:

·         Explain each component of the exploration framework in detail.

·         Provide algorithms to describe the exploration and navigation strategies.

·         Illustrate with clear diagrams/figures.

·         Experiments and Evaluation.

·         Describe simulation setup and performance metrics.

·         Present and analyze quantitative/qualitative results.

·         Compare performance to other methods.

·         Pseudocode and algorithms could be provided as appendices for clarity/reader convenience.

Experiments:

·         Vary parameters systematically in simulated "ablation studies" to evaluate design choices. Describe hardware/software setup in detail for reproducibility.

Results:

·         Present data visually with plots/charts where possible for ease of understanding trends. Perform statistical significance testing of comparisons.

Future Work:

·         Suggest integrating simultaneous localization and mapping (SLAM) to relax reliance on known environment. Discuss enabling real-world deployment with hardware experiments.

Discussion:

·         Analyze how assumptions/limitations impact validity and generalizability of findings. Situate work within broader applications/impact.

Related Publications:

·         Explore journal/conference venues focused on marine robotics, unmanned systems, or artificial intelligence in robotics for dissemination.

Conclusion:

·         Summarize contributions and findings.

·         Discuss limitations and directions for future work.

Author Response

Thank you very much for taking your time to review this manuscript. We really appreciate all your generous comments and suggestions! According to the suggestions and comments, we revised the paper carefully.

Reviewer 2 Report

Comments and Suggestions for Authors

Misuse of the experiment statement is observed. Only simulations are conducted. 

Tables 2-4: misuse of colors in the numbers. 

Tables 4 and 5 are in the conclusion section, which is inappropriate. 

Figures 9-11 are not legible without zooming in.

Other comments:

USV is not self-explanatory in the title.

The abstract is lacking quantitative evidence to support the performance of the framework.

Comments on the Quality of English Language

Moderate editing is required. 

Author Response

Thank you very much for taking your time to review this manuscript. We really appreciate all your generous comments and suggestions! According to the suggestions and comments, we revised the paper carefully.

Comments 1: Misuse of the experiment statement is observed. Only simulations are conducted.

Response 1: We appreciate your valuable comments, which have greatly improved the quality of our paper. Due to our carelessness, we misused the experiment statement. We have made the necessary revisions to clarify that only simulations were conducted and no actual experiments were performed.

Comments 2: Tables 2-4: misuse of colors in the numbers. 

Response 2: Thank you for your feedback. We acknowledge the misuse of colors in the numbers in Tables 2-4. We have corrected this issue to ensure the tables are clear and easy to read.

Comments 3: Tables 4 and 5 are in the conclusion section, which is inappropriate.

Response 3: Thank you for your valuable feedback. We agree that placing Tables 4 and 5 in the conclusion section is inappropriate. We have relocated these tables to the results section, where they are more appropriately discussed and analyzed.

Comments 4: Figures 9-11 are not legible without zooming in.

Response 4: Thank you for your feedback. We understand that Figures 9-11 are not legible without zooming in. We have revised these figures to improve their clarity and readability.

Comments 5: USV is not self-explanatory in the title.

Response 5: Thank you for your feedback. We will clarify the term "USV" in the title by using the full form "unmanned surface vehicle" to ensure it is self-explanatory.

Comments 6: The abstract is lacking quantitative evidence to support the performance of the framework.

Response 6: Thank you for your feedback regarding the lack of quantitative evidence in the abstract. We appreciate your concern. Our method involves extensive comparative data with traditional methods. Including detailed quantitative analysis in the abstract would significantly increase its length, so we opted for a qualitative summary instead. For specific quantitative analysis, please refer to the conclusions in Chapter 5, which covers the results of the simulation experiments in detail.

Reviewer 3 Report

Comments and Suggestions for Authors

The paper presents an exploration method for USV, which is based on a multi-stage exploration strategy and hierarchical navigation. The method was tested in the Gazebo simulator in three different environments, but all under ideal conditions, i.e., no wind, waves, or currents. Even though the authors say that the more realistic simulation is part of future work, it is still not convincing why this has not been done for this publication. Besides the experiments, the contributions of the work are questionable and unclear; the authors claim that they introduced the global and local layers concepts during the exploration process; however, this has already been a known concept. 

1) L. Schmid, V. Reijgwart, L. Ott, J. Nieto, R. Siegwart and C. Cadena, "A Unified Approach for Autonomous Volumetric Exploration of Large Scale Environments Under Severe Odometry Drift," in IEEE Robotics and Automation Letters, vol. 6, no. 3, pp. 4504-4511, July 2021, doi: 10.1109/LRA.2021.3068954.

Also, the authors claim they introduced the hierarchical navigation strategy, but this is also a known concept. 

2) Cai J, Yan F, Shi Y, Zhang M, Guo L. Autonomous robot navigation based on a hierarchical cognitive model. Robotica. 2023;41(2):690-712. doi:10.1017/S0263574722001539

3) Chang L, Shan L, Zhang W, Dai Y. Hierarchical multi-robot navigation and formation in unknown environments via deep reinforcement learning and distributed optimization. Robotics and Computer-Integrated Manufacturing. 2023;83:102570. doi:10.1016/j.rcim.2023.102570

Furthermore, there are statements in the paper which are not entirely correct, such as:

"Frontier-based approaches typically employ image processing to identify boundary data and carry out boundary detection on the entire global map following each map update."

While image processing can be a component of frontier-based exploration, it is not the typical or primary method used but occupancy grids. 

Figure 1 claims that it illustrates the authors' exploration flow relationship, but it is a typical exploration flow.

The authors claim that their method can be adapted to relatively high-speed navigation after subsequent improvement, but they don't elaborate on which improvement. 

 

Additionally, the paper has many typos, some of them are:

Line 83: The first word should be capital.

Figure 1 caption is missing the word 'our'.

Line 105: The sentence is missing the verb.

Lines 399, 403, 404: The sentences seem wrongly connected (a comma should be a dot, and vice versa). 

Line 519: USV should be capital.

Line 583: The citation for Gazebo is not correct. 

Lines 135, 136, 145: The sentence starts with []. 

Author Response

Thank you very much for taking your time to review this manuscript. We really appreciate all your generous comments and suggestions! According to the suggestions and comments, we revised the paper carefully.

1.Ideal Conditions of the Experiment:

We appreciate your feedback regarding the ideal conditions under which our experiments were conducted. Significant interference can indeed impact positioning and mapping. Our current study did not incorporate an anti-interference module, resulting in a "ghosting" phenomenon, which is the drift of map mapping when exploring the sea surface with white noise. Compared to land mobile robots, anti-interference in marine environments presents greater challenges. Addressing this will be a key focus in our future work.

2.Global and Local Layers and Hierarchical Navigation Strategies:

While the concepts of global and local layers and hierarchical navigation strategies are established, our work offers improvements in the allocation and search of known and unknown viewpoints within these layers. Specifically, we have eliminated the global layer concept during exploration. Instead, we focus solely on the local layer for viewpoint allocation. In the upper half of the local layer, we use a dynamically pruned RRT (Rapidly-Exploring Random Tree) to explore unknown areas. This algorithm is more lightweight and optimized, reducing computing resource usage compared to traditional RRT. In the lower half of the local layer, a frontier-based boundary detection method is applied. This approach is more suitable for USVs, as their movement should be more coherent, avoiding frequent direction changes. Thus, we extract rear viewpoints using a simplified method. Additionally, our hierarchical navigation strategy is tailored for unmanned ships, providing more forward-looking exploration by pre-allocating global viewpoints.

3.Clarification on Boundary Detection:

We agree with the reviewer's comment regarding boundary detection. Typically, boundary identification using image processing is a branch of frontier-based exploration. In our practical exploration process, we use the Canny algorithm to analyze and extract boundary points from the occupancy grid, which is a common method. In this regard, we believe our approach aligns with the reviewer's perspective.

4.About typos:

Thank you for pointing out the typos. We have corrected them in the article. We appreciate your careful review and valuable comments.

Reviewer 4 Report

Comments and Suggestions for Authors

An important study for search strategy optimisation when USVs and UUVs are exploring, giving very good results compared to three baselines. The background is often awkward, partly because it starts from too far back in history and then is too abbreviated. Unique aspects of USVs and UUVs compared to say UAVs should be addressed more directly, such as underactuation. The comparison methodology and metrics should be covered before presenting results. There is limited discussion and it is mixed in among the results rather than a distinct section. I enclose some embedded comments as I reviewed.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Author Response

Thank you for your detailed feedback and embedded comments. We have made the necessary changes to the article based on your suggestions and have updated the manuscript accordingly. The modifications have been highlighted in green for your convenience. We appreciate your thorough review and valuable input.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

After a careful revision, I find the manuscript suitable for publication in its present form in the Journal of Marine Science and Engineering without any further changes. The research develops an innovative exploration framework with significant potential to advance autonomous operations of unmanned surface vehicles. The manuscript is clearly written and scientifically robust.

Author Response

Thank you very much for your positive feedback and recommendation for publication. We are thrilled that you find our manuscript suitable for publication in its present form. We are also pleased that you found the manuscript to be clearly written and scientifically robust.

We would like to extend our gratitude for your time and effort in reviewing our work. Your feedback has been invaluable in helping us improve the quality of our manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

For scientific rigorousness as a journal paper, the authors may consider adding scales (rulers) to most figures, even if they are for illustration purposes only.

Again, the usage of the term "simulation experiment" may mislead readers even if simulation is added before the term "experiment". The single term "simulation" suffices to describe the work done. 

A related concern is regarding the claim of optimization contribution. Rigorously, optimization is done based on iterative experiments. Without real experiments in the manuscript, the optimizing work should be clarified as simulation-based optimization.

The authors are advised to justify the statement of "novel" by reviewing other state-of-the-art frameworks. The current manuscript lacks sufficient evidence to support the novelty.

Comments on the Quality of English Language

Moderate editing is required.

Author Response

Comments 1: For scientific rigorousness as a journal paper, the authors may consider adding scales (rulers) to most figures, even if they are for illustration purposes only.

Response 1: Thank you for your suggestion to add scales to the figures. We agree that in certain cases, scales can enhance the scientific rigor of the illustrations, which is why we have already included them in some of the key figures where precise measurements are necessary. However, for the remaining figures, which are intended primarily for conceptual illustration, we believe that adding scales might not contribute significantly to the reader's understanding and could potentially complicate the visual presentation.We hope this explanation addresses your concern. If you have any specific figures in mind where you believe scales would be particularly beneficial, we are open to further discussion.

Comments 2: The usage of the term "simulation experiment" may mislead readers even if simulation is added before the term "experiment". The single term "simulation" suffices to describe the work done. 

Response 2: Thank you for your insightful suggestion regarding the use of the term "simulation experiment." We agree that clarity is essential, and we have revised the manuscript to use the single term "simulation" throughout the text to avoid any potential misunderstanding.We appreciate your careful review and valuable feedback, which have helped improve the clarity of our work.

Comments 3: A related concern is regarding the claim of optimization contribution. Rigorously, optimization is done based on iterative experiments. Without real experiments in the manuscript, the optimizing work should be clarified as simulation-based optimization.

Response 3: Thank you for your thoughtful feedback regarding the need to clarify that our optimization work is simulation-based. In response, we have not only made this explicit in the abstract but have also reiterated it in the conclusion section. These adjustments ensure that the scope and nature of our work are clear throughout the manuscript.We appreciate your insights, which have contributed to improving the clarity and accuracy of our paper.

Comments 4: The authors are advised to justify the statement of "novel" by reviewing other state-of-the-art frameworks. The current manuscript lacks sufficient evidence to support the novelty.

Response 4: Thank you for your feedback regarding the use of the term "novel" in our title. After careful consideration, we have revised the title to better reflect the content and contributions of our work. The new title is: "An Efficient Autonomous Exploration Framework for Unmanned Surface Vehicles in Unknown Waters"We believe this title more accurately represents the focus and significance of our research.We appreciate your valuable suggestions, which have helped us improve the clarity and accuracy of our manuscript.

 

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed most of the questions; however the contributions of the work are still questionable, therefore my overall recommendation is the same. 

Author Response

Thank you for your continued feedback on our manuscript. We appreciate your acknowledgment that most of the questions have been addressed.

Regarding the concerns about the contributions of our work, we understand the importance of clearly demonstrating the significance and novelty of our research. To further clarify and strengthen this aspect, we have revisited the manuscript and made additional revisions to explicitly highlight how our proposed framework advances the current state of the art.

We appreciate your thoughtful review and are committed to improving the quality and clarity of our manuscript.

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for your very thorough treatment of my suggested changes. I now recommend suitable for publication.

Author Response

Thank you very much for your thorough review and for recommending our manuscript for publication. We greatly appreciate your detailed feedback, which has significantly contributed to improving the quality of our work. We're pleased that the revisions meet your expectations.

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

Tables 2,3,4 - the unit of seconds should be added to each column.

Check the figures' captions. There are a few overlaps of meaning, e.g.,

Figure 7. Simulated simulation environment

Figure 11. HBE simulation results. Simulation results of HBE.

 

 

Comments on the Quality of English Language

Moderate editing is required.

Author Response

Thank you for your detailed observations.

1. Tables 2, 3, 4: We have added the unit of seconds to each relevant column to ensure clarity and consistency.

2. Figures' Captions: We have carefully reviewed and revised the captions to avoid any overlap in meaning. Specifically:

   • The caption for Figure 7 has been corrected to remove the redundancy.
   • The caption for Figure 11 has been streamlined for clarity and to avoid repetition.

These revisions have been implemented in the manuscript, and we appreciate your attention to these details.