Dynamic Task Allocation of Multiple UAVs Based on Improved A-QCDPSO

Round 1

Reviewer 1 Report

This paper presents a market auction mechanism for the dynamic allocation of UAV tasks using a discrete particle swarm algorithm with particle quality clustering. The paper is well written.

In the problem formulation part, it is suggested to add an illustration of the target problem for better following the descriptions afterward.

Is the task allocation is a synchronized process? Is PSO applied independently for each UAV or agent? Is the generated algorithm shared by all UAVs, or do they have their own learning mechanism? Is there any central coordinator or is everything distributed? In practice, UAV vision or sensor ranges may not cover the entire environment. How they can deal with a partially observable environment. These informations should be stated clearly in the problem formulation part.

The dynamics of the UAVs are ignored? Can they fly unlimited time? After setting a target or job, how UAV accomplish it should be described. What happens if a task takes extra time due to a change in opponents. These facts are not included within the algorithm. Therefore, the settings need revision for a more realistic scenario.

The proposed method is found to marginally improve the computation time. The significance of such improvement should be discussed clearly?  

Some ambiguous sentences should be revised. For example,

“the requirement of UAV group to the task execution time is getting higher and higher,” Make the statement clear with appropriate causes and results.

“An efficiency function based on the efficiency maximization rule of matrix full permutation is proposed”

Author Response

Thank you very much for your suggestion. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Authors try to introduce the market auction mechanism to design a discrete particle swarm algorithm based on particle quality clustering by a hybrid architecture. The particle subpopulations are dynamically divided based on particle quality, which changes the topology of the algorithm. However there are some comments must be addressed to improve paper 

1- authors should highlight the proposed technique in abstract and introduction with describ the details with comparing to previous work

2- authors should create one section for related work with considering to discuss the applications Predictive estimation of the optimal signal strength from unmanned aerial vehicle over internet of things using ANN,Performance optimization of tethered balloon technology for public safety and emergency communications,Collaboration of UAV and HetNet for better QoS: a comparative study,Multi‐UAV and SAR collaboration model for disaster management in B5G networks,Energy-efficient tethered UAV deployment in B5G for smart environments and disaster recovery. 

A- authors must used more recent work to create table to compare the current work with the previous work with considering Distributed clustering for user devices under unmanned aerial vehicle coverage area during disaster recovery,Multi-Drone Edge Intelligence and SAR Smart Wearable Devices for Emergen,cy Communication,Drones’ Edge Intelligence over Smart Environments in B5G: Blockchain and Federated Learning Synergy,Green IoT for eco-friendly and sustainable smart cities: future directions and opportunities, Bridging the Urban-Rural Connectivity Gap through Intelligent Space, Air, and Ground Networks.

4- authors should explain figure 5 with more details 

5- explain why fitness goes down when iteration is 50, in fig.6 and 8

6-authors should highlight the future work of this work in conclusion by two lines at the end. 

Author Response

Thank you very much for your suggestion. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed my comments and revised them accordingly.