UAV Swarm Scheduling Method for Remote Sensing Observations during Emergency Scenarios

Round 1

Reviewer 1 Report

The authors use interrogative sentences, which is unacceptable for scientific style. " (1) how do we describe the  constraint relationship between the flight range of the task and  the flight capability of the UAV? (2) How can we determine the  constraint relationship between the total flight range of the task  set and the flight capability of the UAV swarm? (3) If each task in 
the task set does not exceed the UAV flight capability and the  total flight range of the task set conforms to the UAV swarm  flight capability, each UAV can be paired with each task such that  we can fully utilize the flight capability of each UAV. "The rest of the paper is written quite well. The method is presented clearly and logically. A large number of experimental results are presented. Perhaps it would be good to draw an analogy with similar methods. To date, there are a large number of similar methods, it is necessary to show the advantage of the presented method.

Author Response

Comments to the Author

Q1: The authors use interrogative sentences, which is unacceptable for the scientific style. " (1) how do we describe the constraint relationship between the flight range of the task and the flight capability of the UAV? (2) How can we determine the constraint relationship between the total flight range of the task set and the flight capability of the UAV swarm? (3) If each task in the task set does not exceed the UAV flight capability and the total flight range of the task set conforms to the UAV swarm flight capability, each UAV can be paired with each task such that we can fully utilize the flight capability of each UAV. "The rest of the paper is written quite well. The method is presented clearly and logically. A large number of experimental results are presented. Perhaps it would be good to draw an analogy with similar methods. To date, there are a large number of similar methods, it is necessary to show the advantage of the presented method.

A1: Thank you for your suggestions. We have modified the three interrogative sentences to become declarative sentences, as follows:

“To facilitate the modeling and solution of the scheduling problem, we decomposed the complex problem into the following three sub-problems: establishing the constraint relationship between the flight range of the task and the flight capability of the UAV; establishing the constraint relationship between the total flight range of the task set and the flight capability of the UAV swarm; modeling the task allocation of UAV swarm and task set, so to fully utilize the operational capabilities of each UAV. ”

According to the suggestions, we added a comparison test to verify the advantages of the proposed method. Since the remote sensing observation during emergency scenarios is very complex, and there is no UAV swarm scheduling or task allocation method that can be directly used, we chose to reduce the complexity of the experiments and thus accommodate the comparison method. We used the direct allocation method as the comparison method, which directly pairings the UAVs and tasks without considering the decomposition of the task set, etc. Since the direct allocation method cannot solve the problems of transit flight range and multiple executions, we removed the constraints of transit flight range and the UAV takeoff and landing points, and used only the flight range of the tasks as the only constraint, while expanding the number of UAVs to avoid the problem of multiple executions. In this case, we conducted 10 tests, and the results show that the convergence time (the time to get the solution scheme) of the proposed method is 93.43% less than the direct allocation method, which verifies the efficiency of the method in this paper. The details can be found in section 4.6.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors,

I found the article interesting, however, a few changes and suggestions shall be taken into account to improve the quality and presentation of the proposed article:

1) Please format all the tables presented in the article. They are not in scientific format and are going out of bounds. Table 3 is going way out of the bounds. Please fix.

Tables 5 also needs to be reformatted. Please carefully reformat all the tables.

 

2) Literature review can be improved. The following works have in detail worked on the UAV operational efficiency and collaborative efficiency, I do not see any such work in the literature review:

a) "Energy-efficient navigation of an autonomous swarm with adaptive consciousness"

b) "Multi-robot coordination for jams in congested systems"

c) "Swarm formation morphing for congestion-aware collision avoidance"

The proposed work can actually be at least naively compared with the article "c)"  to show the efficiency of the proposed method.

 

3) Motivation for the proposed methodology can be highlighted for the readers.

 

4) Equations need to be properly formatted. Furthermore, the equation numbers should be right-aligned with the equation itself left-aligned. Eq. 12 & 13 look very unprofessional and messy. 

 

5) Conclusions can be improved. A clear future direction is missing. Please write a separate clear paragraph highlighting the future directions the authors see this work can be extended for further contributions in the research community.

 

Thanks

Author Response

Q1: Please format all the tables presented in the article. They are not in scientific format and are going out of bounds. Table 3 is going way out of the bounds. Please fix. Table 5 also needs to be reformatted. Please carefully reformat all the tables.

A1: Thank you for your suggestions. We have reformatted table 3, table 5, and other tables in the paper, so that they are not going way out of the bounds.

Q2: Literature review can be improved. The following works have in detail worked on the UAV operational efficiency and collaborative efficiency; I do not see any such work in the literature review:

1. a) "Energy-efficient navigation of an autonomous swarm with adaptive consciousness"
2. b) "Multi-robot coordination for jams in congested systems"
3. c) "Swarm formation morphing for congestion-aware collision avoidance"

The proposed work can actually be at least naively compared with the article "c)" to show the efficiency of the proposed method.

A2: Thank you for your suggestions. We have updated the references in the introduction and supplemented the references "a" and "c" you mentioned, as follows:

“Currently, the UAV swarm scheduling method is mainly used for military reconnaissance scenarios, emphasizing UAV obstacle avoidance, target identification, and strikes, among others. For example, Jawad et al. proposed a method to reduce the unnecessary power consumption of sensors and improve the overall energy consumption and a congestion control method based on thin-plate splines technique for maintaining formation and avoiding collisions of UAV swarms, respectively [32-33].”

Since this paper mainly involves the application of the UAV swarm and has a weak correlation with the reference "b)", we will refer to it in the following research. Meanwhile, “the optimal distribution of a swarm formation on either side of an observer methodology” proposed in article “C” is very innovative and has important implications for the formation retention and efficiency of the UAV swarm. We compare the two papers at the literature level in the penultimate paragraph of the introduction section, “Similar to the objective of reference [33], it is to improve the efficiency of UAV swarm.”, and since the application scenarios of the two paper are slightly different, we are going to compare them at the method level in the subsequent study. As a supplement, we have made simple comparison with the direct allocation method, which can be found in section 4.6.

Q3: Motivation for the proposed methodology can be highlighted for the readers.

A3: This is a very constructive suggestion. We have highlighted the motivation of the proposed method in the abstract and introduction section. In particular, we have added a paragraph in the introduction to emphasize the motivation for writing this paper, as follows:

“In 2020, flooding occurred in Poyang Lake, China, and the disaster caused many villages to be flooded, and people's lives and properties were seriously threatened. We were responsible for the UAV remote sensing observation of dozens of discrete affected villages. We used multiple UAVs to acquire remote sensing data from the affected villages by manual scheduling method. However, we found that this method was inefficient and limited the operational capabilities of the UAVs. We reviewed a lot of literature and did not find a suitable UAV swarm scheduling method that could meet the remote sensing observation requirements of emergency scenarios, so we proposed the UAV swarm scheduling method that is applicable to remote sensing observation of emergency scenarios from the real demand. Similar to the objective of reference [33], it is to improve the efficiency of UAV swarm.”

Q4: Equations need to be properly formatted. Furthermore, the equation numbers should be right-aligned with the equation itself left-aligned. Eq. 12 & 13 look very unprofessional and messy.

A4: Thank you for your suggestions. We have formatted Eq. 12 & 13, and all the rest of the equations, and they are now clearer and aligned.

Q5: Conclusions can be improved. A clear future direction is missing. Please write a separate clear paragraph highlighting the future directions the authors see this work can be extended for further contributions in the research community.

A5: The suggestion is taken. According to the suggestion, we used a paragraph to state the future research, as follows:

“Further research includes: the selection of UAV landing and takeoff points should consider the influence of roads, i.e., the scheduling method should add road constraints; consider the route planning of mission areas and the risk posed by the collision probability of multiple UAVs in adjacent task areas during flight; consider increasing the temporal constraints of missions from mission formulation; the scheduling method also need to consider compatibility with different data types, different sensors, and different UAVs.”

Author Response File: Author Response.docx

Reviewer 3 Report

The authors address the issue of remote reconnaissance using unmanned aerial vehicles (UAVs) in emergency security scenarios. The current trend in remote sensing using multiple UAVs has so far been characterized by high manpower costs and limited capabilities. 
 To address the multi-resource challenges, this study contributes and proposes a swarm of UAVs and 
planning method. Considering the observation of remote sensing of the affected sites
at Poyang Lake in China, the authors tested the effectiveness of the proposed method described as an example. 
The results showed that the average task completion time for 5 UAVs controlled by the crew using the UAV swarm planning method was 7.23 h. The flight range utilization rate for each UAV was 90.42%. the transmission flight range was 95.92 km and the number of operational flights for each flight was 40.
The average task completion time for each UAV using manual 
was 7.34 h. The range utilization rate for each UAV was 60.18%, the transfer range was 115.44 km, and the number of transfer flights was 1.8%. 
The operational flights were 62. 
The results showed that the UAV swarm planning is more effective better than the manual planning method, especially in terms of the performance of significantly reducing the cost y of manpower. 

The work is based on scientific procedures and the comparison tests conducted proves the improvement in efficiency and other parameters of UAVs and for the proposed/described method. It contains elements of novelty.

Author Response

Q1: The authors address the issue of remote reconnaissance using unmanned aerial vehicles (UAVs) in emergency security scenarios. The current trend in remote sensing using multiple UAVs has so far been characterized by high manpower costs and limited capabilities. To address the multi-resource challenges, this study contributes and proposes a swarm of UAVs and a planning methods. Considering the observation of remote sensing of the affected sites at Poyang Lake in China, the authors tested the effectiveness of the proposed method described as an example. The results showed that the average task completion time for 5 UAVs controlled by the crew using the UAV swarm planning method was 7.23 h. The flight range utilization rate for each UAV was 90.42%. the transmission flight range was 95.92 km and the number of operational flights for each flight was 40. The average task completion time for each UAV using manual was 7.34 h. The range utilization rate for each UAV was 60.18%, the transfer range was 115.44 km, and the number of transfer flights was 1.8%. The operational flights were 62. The results showed that the UAV swarm planning is more effective better than the manual planning method, especially in terms of the performance of significantly reducing the cost y of manpower. The work is based on scientific procedures and the comparison tests conducted to prove the improvement in efficiency and other parameters of UAVs and for the proposed/described method. It contains elements of novelty.

A1: Thank you for your recognition of the innovation of this paper. With the suggestions of other reviewers, we have revised and improved the article. I hope you can continue to track our manuscript.

Author Response File: Author Response.docx

Reviewer 4 Report

Aiming at the problems of insufficient network and low operation efficiency of multiple UAVs in emergency remote sensing observation, this paper proposes a UAV swarm scheduling method based on centralized network architecture. However，there are some problems in the paper as follows:

1)The innovation is deficiency. The key algorithm in this paper uses the existing particle swarm optimization algorithm. In addition, the equilibrium constrained K-mean algorithm and the location acquisition algorithm are existing algorithms；

2)The results are not intuitive. The evaluation indicators are not explained, and it is not clear how to evaluate the experimental results；

3)There is a problem with the formula layout. Equations 12,13 and 25 are both problematic；

4)The title of the figures should be clear and concise.The explanation of the figures should be above or below them, not in the title of the figures.

5)There are few new reference of recent 3 years.

Author Response

Q1: The innovation is deficiency. The key algorithm in this paper uses the existing particle swarm optimization algorithm. In addition, the equilibrium constrained K-mean algorithm and the location acquisition algorithm are existing algorithms.

A1: Thank you for your comments. We know that remote sensing observation during emergency scenarios with the UAV swarm is a new problem with the development of UAV remote sensing. How to fully utilize the operational capabilities of the UAV swarm and complete the tasks more efficiently is the main emphasis of current research. The existing UAV swarm scheduling methods are mainly focused on military reconnaissance and strike, which is completely different from the remote sensing observation tasks. The existing UAV swarm scheduling methods cannot solve the problem of remote sensing observation in multiple discrete areas (the different locations and sizes of the tasks need to be considered). Therefore, this paper develops a scheduling method for remote sensing observation during emergency scenarios.

The purpose of this paper is not to innovate algorithms, but to solve new problems that have not yet been resolved. Several existing algorithms are used in this method, it may be considered as an innovative way to synthesize the existing algorithms so as to solve new problems. If we do not analyze the specific application scenarios, clarify the needs and limitations of the scenarios, and formulate corresponding solutions, these existing algorithms are not helpful for practical application. This article originates from a practical problem we encountered during the flood monitoring of Poyang Lake in 2020, the scenario we faced was that we needed to acquire data from dozens of affected villages with several UAVs, we chose the manual scheduling method and found that the efficiency of UAVs could not be fully utilized. We reviewed a large number of kinds of literature but found no suitable method that could be used directly, and the scheduling methods of UAV swarm in the military field were not applicable. We started from real needs, analyzed the problems, and then extracted solution ideas, and finally form this paper.

The innovation of this paper is mainly to propose a comprehensive scheduling method: firstly establishes the constraint relationship between remote sensing task and a UAV, and solves the problem that the operational flight range of a task exceeds the maximum flight range of a UAV; then establishes the constraint relationship between the taskset and UAV swarm, and solves the problem that the operational flight range of the taskset exceeds the maximum operational capacity of UAV swarm; finally, on the basis of satisfying the above two problems, the problem of pairing the UAV swarm with the tasks within the task subset is solved. After comparing the proposed method with the direct allocation and manual scheduling methods, the results show that this proposed method can solve the problem of efficient scheduling of UAV swarm during emergency scenarios, and also outperforms the direct allocation method in terms of time efficiency. Among them, the direct allocation method is performed in an ideal situation with many limitations of the actual scenario removed, and cannot solve the practical problems.

The above is our innovative description of the article.

Q2: The results are not intuitive. The evaluation indicators are not explained, and it is not clear how to evaluate the experimental results.

A2: Thank you for your suggestions. We reframed the results, expressed each evaluation indicator, explained their meaning, and how to reflect the efficiency and effectiveness of the proposed method. The results contain 6 sections, of which the important comparison results include 3 sections, which are “4.3 Task set decomposition results”, “4.5 Comparison results with manual scheduling method” and “4.6 Comparison results with manual scheduling method”.

In the 4.3 section, we added a description of how to use variance to evaluate the effect of task subset decomposition, as follows:

“Variance is an indicator of the difference operation flight in different task subsets. A small value of variance indicates that the task-balanced restricted K-means can well balance the operation flight in different task subsets, which can ensure that the operational capabilities of UAV swarm can be fully exploited.”

In the 4.5 section, we added descriptions of “the operational flight range”, “the transit flight range”, “total flight range”, “total sorties”, “flight range” and “the average utilization rate of UAV”, et al., as follows:

“The operational flight range is the sum of the operational ranges of all tasks in the task set, which is 1206.1km; The transit flight range is the distances between the different tasks and between the tasks and UAV takeoff and landing points. The smaller the value, the higher the efficiency. It can be seen that the scheduling method is 95.92km, which saves 16.97% compared with 115.44 of the manual scheduling method; Total flight range is the sum of operational flight range and transit flight range; Total sorties refer to the total number of UAV flights required to complete the task set. It can be seen that the total number of sorties of the scheduling method is 40, which saves 34.43% compared with the manual scheduling method; Flight range refers to the range that the total sorties; The average utilization rate of UAV refers to the real total flight range divided by the flight range that able to work, which reflects the operational capabilities of a UAV. It can be seen that the algorithm in this paper is 0.9, which is 50% higher than that of the manual scheduling method; It can be seen that the scheduling method is better than the manual scheduling method in transit flight range, total sorties, and average UAV utilization, which shows the usability of the scheduling method.”

Section 4.6 is a new comparative test based on the suggestions of reviewers 1 and 2. It is to compare the proposed method with the direct allocation method, mainly evaluated the solution efficiency of the schemes, and use the convergence time to evaluate the efficiency of the two methods. The shorter the convergence time, the higher the efficiency of the method. The details are as follows:

“The two methods are tested 10 times, and the test results are shown in Table 6. It can be seen that the shortest convergence time of the direct allocation method is 20.1s, and the average convergence time is 27.88s; while the shortest average convergence time of the scheduling method is 1.11s, and the average convergence time of the 10 test is 1.32s. Since the task subsets are operated in parallel, the average convergence time of the task set is calculated. The time required for task set decomposition and the selection of takeoff and landing points is relatively short, so the time is not calculated. It can be seen that, compared with the direct allocation method, the scheduling method shortens the computation time by 93.43% under the condition that the task complexity is reduced for adapting to the direct allocation method. The results show that the scheduling method is more efficient compared to the direct allocation method.”

Q3: There is a problem with the formula layout. Equations 12,13 and 25 are both problematic；

A3: Thank you for your suggestions. We have revised equations 12, 13, and 25.

Q4: The title of the figures should be clear and concise. The explanation of the figures should be above or below them, not in the title of the figures.

A4: This is a very good suggestion. The title of the figures and tables have been revised and the explanation has been put into the text to make it more concise.

Q5: There are few new references of recent 3 years.

A5: Thank you for your suggestion. We have updated the references.

Author Response File: Author Response.docx

Reviewer 5 Report

I believe this paper is relevant and makes a contribution but it has to be better presented before being able to accept it for publication.

The main problem is with equations. for example, equations 13 and 25 are not clear and require editing to be read and understood. In general, the mathematical notation needs improving and it has to b taken to a more appropriate standard. Please use proper mathematical notation and a notation editor to typeset the equations. These should be not only reported but discussed in the text.

Also the tables with results need some editing they exceed the margins.

You may want to improve upon the literature review with more kinds of studies employing drones e.g. in the context of sustainable agriculture as done in 'Oil palm detection via deep transfer learning'. by Isis Bonet et al., etc.

Author Response

Q1: The main problem is with equations. for example, equations 13 and 25 are not clear and require editing to be read and understood. In general, the mathematical notation needs improving and it has to be taken to a more appropriate standard. Please use proper mathematical notation and a notation editor to typeset the equations. These should be not only reported but discussed in the text. Also the tables with results need some editing they exceed the margins. You may want to improve upon the literature review with more kinds of studies employing drones e.g. in the context of sustainable agriculture as done in 'Oil palm detection via deep transfer learning'. by Isis Bonet et al., etc.

A1: Thank you for your suggestion. We have revised all the equations such as equation12, 13, and 25, etc., and also explained the purpose and idea of the construction of the equations in the article.

For example:

(1) We interpret the weighting factor in equation (2), “The operational flight range of the task is decomposed by the maximum flight range of the UAV and the weighting factor, and the weighting factor is set to take into account the range consumption of takeoff and landing.”

(2) We explain the purpose of task set decomposition, “The purpose of task set decomposition is to decompose and execute the task set multiple times when the total operational flight range of the task set exceeds the maximum flight range of the UAV swarm.”

(3) We discuss the number of subsets of tasks, “Firstly, the number of task subsets was determined for the clustering of task subsets.”

(4) We discuss the choice of the UAV take-off and landing points, “The purpose of UAV takeoff and landing point selection is to ensure the minimization of the sum of the total transit flight range, so as to reduce the unnecessary consumption.”

We have also corrected all the tables and they are now not out of bounds. We have updated the references to include the literature you suggested, as follows:

“UAV remote sensing has been widely used in land surveys, agriculture, forestry, disaster rescue, national security, and many other fields [9-11].” The literature [9] is “Oil palm detection via deep transfer learning'. by Isis Bonet et al., etc.”

Author Response File: Author Response.docx

Round 2

Reviewer 4 Report

Currently, unmanned aerial vehicle (UAV) remote sensing mode has the defects of high labor cost and limited operation ability. In addition, there is no UAV swarm scheduling method applicable to remote sensing emergency scenarios. In order to resolve these problems, an effective UAV swarm scheduling method is proposed in this paper. At present, there are the following problems:

• There is something wrong with the layout. The title of Figure 3 should be on the same page as the picture, and the title and table of Table 4 should be on the same page;
• The interpretation of in Formula 16 and 17 would be “is the total operational flight range of the UAV j”;
• Figure 9 is not clear;
• The width of Table 6 exceeds the width of the page and should be adjusted.

Author Response

Q1: Currently, unmanned aerial vehicle (UAV) remote sensing mode has the defects of high labor cost and limited operation ability. In addition, there is no UAV swarm scheduling method applicable to remote sensing emergency scenarios. In order to resolve these problems, an effective UAV swarm scheduling method is proposed in this paper. At present, there are the following problems:

(1) There is something wrong with the layout. The title of Figure 3 should be on the same page as the picture, and the title and table of Table 4 should be on the same page;

(2) The interpretation of in Formula 16 and 17 would be “is the total operational flight range of the UAV j”;

(3) Figure 9 is not clear;

(4) The width of Table 6 exceeds the width of the page and should be adjusted.

A1: Thank you for your suggestions. We have revised the above four problems as follows:

(1) We have placed the title and picture of Figure 3 on the same page, and the title and table of Table 4 on the same page.

(2) We have changed “ is the total operational flight range of the UAV” to “ is the total operational flight range of the UAV” in the interpretation of Formula 16 and 17.

(3) We redrew Figure 9, increasing the size of the fonts and making them bold. The curve color was also changed to black to make it clearer.

(4) We have adjusted the width of Table 6, and it now does not exceed the page.

 

Your suggestions have improved the quality of our manuscript. Thanks again.

Author Response File: Author Response.docx