A Novel Multi-Robot Task Assignment Scheme Based on a Multi-Angle K-Means Clustering Algorithm and a Two-Stage Load-Balancing Strategy
Round 1
Reviewer 1 Report
A practical task assignment is one of the core problems of a multi-robot system. In this paper, a multi-robot task assignment strategy based on load balancing is proposed to effectively balance and plan out the execution cost of each robot when it has a large number of working task points. The proposed method can well solve the uneven execution cost of each robot in the task assignment process, and effectively improve the efficiency of the system in completing tasks
Author Response
Response: Thank you for your affirmation to our manuscript.
Reviewer 2 Report
The topic of the paper is Multi-robot Task Assignment. The task on which the authors demonstrated their system is based on the processing of a series of tasks in space by a group of robots. This task is similar to the Travelling Salesman Problem (TSP), with the addition that at each node the robots have to perform some work (product storage and selection). The authors combine several known and original methods. They use a modified K-Means algorithm to partition the task points to clusters, then the original algorithm for adjusting, and then a genetic algorithm is used to solve the route-finding problem for each robot.
I have the following criticisms of the text:
In my opinion, the mathematical model has shortcomings, some of which I mention below
- Formally, for relation (2) I have an idea what the authors want to express, but I believe that it is not expressed correctly in this way
- The storage and retrieval model described from line 164 onwards is not well argued and it is not clear why it is developed in this way. Moreover, it is not clear to me how this can affect the system described
- Relations (14) and (18) use indices that I don't understand very well. In my opinion, the indices on the left side of the equation and in the numerator should be the same, and in the denominator a different index is used for the sum
- They denote the robot, and the robot is defined by route (150), which I don't think is very appropriate. Moreover, the symbol R is used again as points at some distance from the centre (346)
I would also like to state my other objections:
Many approaches are used to solve TSP, not only genetic algorithms (GA). Therefore, this could have been mentioned in the text, since to my knowledge GAs do not always guarantee the best results
On line 281 it states that the result is the optimal solution after using adjustment strategies. But I am not completely convinced by the text that solving such an output is guaranteed to be optimal.
The evaluation of the results is weaker. The evaluation method uses the classic K-means to demonstrate that the version of this algorithm used improves the results slightly, on the order of tenths of a percent. It is further shown that the support phase solutions achieved by the looping algorithm used further improve. However, it would be appropriate and desirable for the authors to indicate how their solution holds up against the current best approaches for solving the problem.
The text needs to be checked especially in the parts that use the formal approach and recheck at least what I have stated above. A better argument for using the stated approach to advance the current state of knowledge on how to solve multirobot problems would greatly improve the usefulness of this text.
Author Response
Thank you for your comments concerning our manuscript entitled “A novel multi-robot task assignment scheme based on a multi-angle K-means clustering algorithm and a two-stage load-balancing strategy” (ID: electronics-2573191). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made corrections which we hope meet with approval. I have revised the manuscript in a revised state, and the changes are shown in the manuscript. We have added new fund information, which has been shaded in yellow in the manuscript.
We highlighted the changes to my manuscript within the document by using shaded text (yellow) in the manuscript and blue text in the response comment.
Author Response File: 
Author Response.pdf
Reviewer 3 Report
The problem is well stated and motivated. The proposed solution is clear and well explained.
The manuscript has a clear and comphrensive literature review.
The paper has a sufficient amount of recent references.
The proposed algorithms are well presented, are clear and organized.
Comments for improving the text:
Page 10, line 371: In section 4, I suggest that authors use the term "simulation tests and analysis" rather than "experiments and analysis". The title gives the impression that experimental tests will be presented when in fact they are just numerical simulations. I also suggest authors to specify the value of the sample time used in simulation tests.
Author Response
Response: Thank you for your affirmation to our manuscript.
Comments for improving the text: Page 10, line 371: In section 4, I suggest that authors use the term "simulation tests and analysis" rather than "experiments and analysis". The title gives the impression that experimental tests will be presented when in fact they are just numerical simulations.
Response: Thanks for reviewer’s valuable review. We have changed the title of section 4. Thank you for your detailed comment.
