MCO Plan: Efficient Coverage Mission for Multiple Micro Aerial Vehicles Modeled as Agents

Round 1

Reviewer 1 Report

This work shows the coverage mission plan for multiple MAVs considering the balance of resources. The method is clearly described with plenty of results shown. Nevertheless, this work could be improved in the following aspects:

1. The authors show the topology network in Figure 17, but how to assign the path to each MAV agent? By the figures and tables shown later, it is difficult to like them to the actual flight restuls. 

2. The authors try to show the complete simulation results of each case, but how do these cases reflect the effectiveness or efficiency of the proposed method?  

3. It there any visualization results from the GAMA platform?

4. Figure 12 should be the core contribution of this work, but it is not easy to be understood. Could you explain or draw it more clearly?

 

Author Response

We would like to thank reviewer #1 for his invaluable comments and his insightful suggestions. In the attachment you will find the answers and changes made to the paper for each suggestion.

Author Response File: Author Response.pdf

Reviewer 2 Report

Please look at the below. 

 

Paper Review Report

MCO Plan: Efficient Coverage Mission for Multiple Micro Aerial Vehicles modeled as Agents

Liseth Viviana Campo 1*, Agapito Ledezma 2 and Juan Carlos Corrales 1

A brief summary

• Overview

Master, Coordinator, Operator (MCO) Plan is proposed for balance in resources as well as for efficient deployment of MAV fleet in coverage missions within restricted workspaces.

MCO plan is compared with two other approaches: Boustrophedon decomposition, K-means clustering.

The proposed MCO plan is based on the heterogeneous polygon-based coverage mission plan with Voronoi-tessellation.

Each MAV agent follows the coverage path by Firefly algorithm and interact with neighbors in communication range.

• Methodology (Simulation GAMA platform)

Workspaces for testing the heterogeneity and connectivity are Rejoya Farm, Naranjos Farm, Urban Zone in Cauca, Colombia with Zones of Low Interest.

MAV agent is modeled with Belief-Desire-Intention control. Communication takes place based on Foundation for Intelligent Physical Agents protocols.

MAV agent observes the world; coverages are defined, cells and waypoints based on camera parameters are obtained.

Hierarchical coordination is formed by allocating roles (Master, Coordinator, Operator) to each agent model. Tasks are allocated according to agent roles.

Each agent is allocated with a path based on coverage task calculation. Task allocation adapts the FIPA English Auction Interaction Protocol Specification. Each bid and task consider three features: flight time, decision capacity, and communication skills.

• Results

Better resource management is achieved when MCO plan is used, compared to Boustrophedon-based and K-means clustering-based planning.

Allocated roles and coverage tasks derived from MCO plan achieves the balance between heterogeneity and connectivity.

Heterogeneity pattern effectively influences the removal of the energy consumption of the MAV fleet. Energy consumption can be improved with the coverage paths calculated with the Firefly Algorithm.

 

 

General concept comments

• Overall, the manuscript is relevant for the field. However, some of the points are not well-structured enough for reader’s clear understanding.

• The cited references include recent publications though most of the references cited in section 2. Related Work are dated publications. The paper has self-citation but does not have any excessive self-citation.

• Comparing the proposed idea with two other methods seem appropriate to test the hypothesis.

• Manuscript’s method section needs some fill-ins in detail to allow the reproduction of results.

• The figures/tables could be revised for a better conveyance of the idea.

Line 289 mentions ci and refers to Figure 2., while Figure 2. does not have ci.

Figure 3. outlines the BDI control and Table 1. describes the rules for BDI control. Reader is likely to compare the two visuals.

Both Figure and Table should have the same order in description. ‘Desires’ list in Figure 3. should have ‘Send query’ before ‘Read_msg’ to follow the order in ‘Beliefs’ > ‘See_neighbors’, ‘Receive_msg’.

Table 1. could use highlighted rows to differentiate between cylinders and rectangles in Figure 3.

Table 1. could add another column on the far left to list the order of procedure.

Figure 4. Image 4 has a dot on the upper left corner, though it lacks an explanation of what it indicates.

Figures through 8 to 10 are unclear. For example, line 412 states that the ‘centroids are centered in p0’, but the points p1, p2, p3 do not seem like centroids. Also, line 429 ‘The centroid is moved one-third of radius closer to.’ is unclear since there is no distinct reference in the figure.

Figure 11. raises a question of why the MAV icon third from the bottom is an operator and not a coordinator instead.

Tables 9 to 11 data does not match the numerical analysis in lines 786 to 789. The calculated average values are different. (Table 9., Table 10. MCO, Table 11. MCO)

Overall, the order of workspaces (Rejoya Farm, Naranjos, Urban Zone) differs from figures to tables. It would be helpful in comprehension if the order of workspaces stays consistent.

• The conclusions are logical with the statements presented in the manuscript.

 

 

 

Specific comments

• Unclear if Figure 3. ‘Environment’ > ‘World knowledge’ refers to allocated task in line 334.

• Lines through 338 to 342 explains the Figure 3. Though mixed expressions ‘actions’ and ‘goals’ cause confusion.

• Lines 428-429: the phrase ‘most significant sub-area’ is unclear.

• The phases explained in pages 10 to 12 are vague. Figures need more details for better understanding.

• Line 542: 4.2.5. instead of 4.2.1.

• Line 674: The order of workspaces differs from previous tables and figures’ description. Curious if there is a specific reason to this certain ordering.

• Line 727: In Figure 15. (c), Boustrophedon does not show a decreasing trend.

• Line 779: There is no yellow column in Figures 16(a), 16(c), and 16(e).

• Lines 792-793: In Table 10., MCO plan shows a battery decrease over 30% (63.87% battery remained, thus the battery has decreased, 36.13%).

Author Response

We would like to thank reviewer #2 for his invaluable comments and his insightful suggestions. In the attachment you will find the answers and changes made to the paper for each suggestion.

Author Response File: Author Response.pdf

Reviewer 3 Report

 For design of fault-tolerant cooperative MAVs for large and restricted workspaces, methods of resolving an efficient deployment of the MAVs fleet, considering user expectations, are crutial, limited flight time and energy. In the paper, authors proposed plan which created a hierarchical society with roles defined by adequate flight time, communication skills, and decision capability This plan is based on Master, Coordinator, and Operator roles.

The paper is oryginal, well organized, and should be interesting for the UAV community. I'm just wanerding about the state-of-the-art. Shouldn't be added more newer papers? There are just 4 papers from the year 2020, four from 2021, and three from 2022, in total number of 40 cited papers. 

Please standarize the names of Sections. In some you use capital letters, and e.g. in line 944 is mixed convention: Conclusion and Future work. The same is with the line 252.

In the line 649 "dot" is missing.

Please increase the quality of figures.

Author Response

We would like to thank reviewer #3 for his invaluable comments and his insightful suggestions. In the attachment you will find the answers and changes made to the paper for each suggestion.

Author Response File: Author Response.pdf