Conflict-Free 3D Path Planning for Multi-UAV Based on Jump Point Search and Incremental Update

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. Abstract, P1: The abstract states that "conflict-free path planning time decreased by 28% compared to traditional search and conflict resolution algorithms under high-density scenarios of 120 UAVs per square kilometer." while Conclusion 4) states that "the DOCBS+IJPS framework reduces the total planning time by an average of 35.56% while maintaining comparable optimality in total path cost". This discrepancy requires clarification on the benchmark for the 28% reduction (e.g., whether it refers to a specific sub-module comparison) and alignment of
2. Section 3.2.1, P8: The manuscript mentions that the 3D JPS search direction has been expanded to 26 directions (including body diagonals), but it does not analyze whether this direction setting complies with the drone's motion constraints (such as the maximum horizontal turning angle, maximum climb angle defined in Section 2.2).
3. Section 3.4, P13: Figure 5 does not clearly show the interaction details between the upper-layer DOCBS algorithm and the lower-layer IJPS algorithm. It is recommended to optimize the flowchart. In particular, the connection process between how incremental updates are triggered after conflict detection, constraint tree node expansion, and local path re-planning should be clarified.
4. Section 4.1, P13: The manuscript uses a 16m grid unit scale for the airspace model but lacks detailed justification for this choice. The authors point out that "the grid scale significantly impacts path planning". However, there is a lack of comparative experiments on path optimality and computational efficiency at different grid scales (e.g., 8m and 32m), making it impossible to verify the rationality of the 16m scale. It is recommended that a grid scale sensitivity analysis be supplemented.
5. Section 4.1, P14: Table 1 sets the risk cost weight coefficient α₂=0.2 and the actual cost weight coefficient α₁=0.8, but does not specify the method used to determine these proportions. The weight coefficients directly affect the trade-off between path safety and length. Parameter optimization experiments (such as the impact of different α values on path risk and computational efficiency) should be conducted to verify the rationality of the current settings.
6. Introduction section: a kind suggestion is that authors can introduce and address some recent works on path planning methods for UAVs. For example, Research on dynamic particle swarm optimization for multi-objective reconnaissance task allocation of UAVs, Combinatorial optimization for UAV swarm path planning and task assignment in multi-obstacle battlefield environment.
7. Conclusions, P22: The conclusion mentions that “its real-time dynamic path planning performance in complex and dynamic environments requires further improvement”, but does not discuss the extensibility of the existing framework to dynamic obstacles (such as sudden obstacles and UAV failures). It is recommended to supplement specific directions for future work, such as how to extend the incremental update mechanism to dynamic obstacle scenarios, or introduce predictive models to improve real-time performance.
8. Please correct the formula format issues, such as the blank line in Equation (6).

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. There is no analysis with respect to real cases , all the simulations are just based on static simulation and grid based analysis , what is the reason the authors believe this setting replicate the real cases?
2. No sensitivity analysis of the simulation parameter ans how the results will be changed if the simulatiom setting changed ?
3. Update is happening only if conflict happens but how if local changes happen frequently ? the cumulitive impact of these changes can reduce the network efficiency.
4. Some important constraints are ignored, like battery limitation or the bandwidth constraints in the networks ,... 
5. Figure captions are not enough meaningful , needs to be updated in a clearer way
6. No section discussed the future work of the proposed method , authors need to clearly talk about the limitations of their work and how the future work can improve it 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

I have some thoughts and suggestions to offer regarding the paper “Conflict-free 3D Path Planning for Multi-AV Based on Jump Point Search and Incremantal Updtae”  written by Lu Y. et al. This paper presented a novel method to solve the problem of designing large-scale paths without conflicts in urban low-altitude airspace by utilizing an incremental update mechanism and three-dimensional jump point search.  The authors demonstrated the excellent performance of the integrated solution of 3D JPS and DOCBS throughout simulation experiments. In this regard, the paper's ideas and results can make a significant contribution to the development of related fields. However, there are some major and minor errors in the editing of the manuscript that need to be corrected before its publication, which are written on separate pages.

Best regards,

Reviewer

 

Common comments:                                                                

1. Mentioning the limitations of the research results. 

 The majority of the findings in this study are derived exclusively from simulations. However, these results may differ significantly in real-world physical environments. In particular, the payload utilized in experiments and sudden weather changes, particularly the effects of gusts or crosswinds, can have a significant impact on the drone's flight path. It would be beneficial for readers to include these facts in the paper's introduction or conclusion.

2. Spacing between Text and Figures

 Please give a space between text and figures in Line 236, 299, 357 and so on.

3. Citing Formulas 

 Please provide references for formulas, if they were cited from other literatures. 

 

Additional comments: 

 Line 167: 2.2 Single UAV Path Planning Model

Please rewrite all variables such as xi, yi, ci, etc. as subscripts in the text exactly as they are  written in the formulas.

Line 218: 

Is li(t)  different from the waypoint li(t) defined in Line 218?  If it has a different meaning, it would be better to represent it with a different symbol to avoid confusion.

Line 336: Give a spacing

Please give a spacing before start sub-chapter 3.2.2. 

Line 444: Oblique photogrammetry

Please explain the reason why did you utilize oblique photos? 

Line 452: Title of Z-axis in Figure 6

The title of Z-axis (Z) is missing. 

Line 463: Units in Table 1

Please check the correctness of the units “rad” for maximum turning angle and climb angle.

Line 480: Title of Z-axis in Figure 7

 1) The title of Z-axis (Z) is missing in Figure 7(a). 

 2) The topview does not correspond to the 3D view, so it needs to be explained that this is 

  just a simple example.

Line 533: Changes of Graph in Figure 10

 The changes in the graph in Figure 10 are very constant, so why do you say they show  fluctation?

Line 552: Difficulty in recognizing the qualitative values described in Figure 11

 The percentage value of 49.4% for the high-level conflict resolution and resampling accounting of the total time could not easily be seen numerically from Figure 11. It is also  difficult to calculate about 1.96 seconds to plan a conflict-free path for each UAV. An alternative more convincing visual expression method or a detailed numerical description is required because it is not easy for readers to grasp the numerical values with naked eyes.

Line 556: Recommendation for rewritting

The drone path is represented by semi-transparent grid nodes of different colors, with the red grid representing the conflict nodes. After a conflict occurs, it is necessary to resolve it and replan the path, as shown in Figure 12 (b). The replanned path no longer violates conflict constraints, resulting in a conflict-free route.

Line 578: Please provide the basis for obtaining 2.5%

How did you get the value of 2.5% for a total flight distance difference? 

Line 591: Description of the % values showed in Figure 14. 

  As we can see in Figure 14(a), the consumption time varies depending on the number of drones. Please provide a supplementary explanation on how 25.62% was calculated, as written in Line 588. Please also provide the basis for other % values, as described below Figure 14.

- The End –

Comments on the Quality of English Language

Line 556: Recommendation for rewritting

The drone path is represented by semi-transparent grid nodes of different colors, with the red grid representing the conflict nodes. After a conflict occurs, it is necessary to resolve it and replan the path, as shown in Figure 12 (b). The replanned path no longer violates conflict constraints, resulting in a conflict-free route.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

I am grateful for the kind responses and precise corrections you have provided to my all comments. This manuscript is highly suited for publication in its present format.

Best regards,

Reviewer