Behavior Prediction Based Trust Evaluation for Adaptive Consensus of Quadrotors

Round 1

Reviewer 1 Report

The main idea of this paper is by using behavior prediction to deal with faulty agents in a consensus group. I think the paper is generally well written, but can be improved in the following aspects:

1. I think the title can remove the word "An" before "Adaptive", and the English language can be improved;

2. The motivation and contributions of the paper can be better explained. For example, why don't use similar method to predict malfunction of one quadrotor by itself? why it is necessary to be predicted by other quadrotors, instead of detecting one's fault and broadcast out? 

3. What if the method gives wrong prediction? What the method would perform if disturbances exist? 

 

I think the paper can be better improved by answering those questions.  

Author Response

Dear Reviewer:

Thanks for the valuable comments. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear

The paper presents a quite interesting approach to the problem. Unfortunately The mathematical aspects of that solution are beyond a scope of my area o interest, so I am not able to evaluate a theoretical side of the paper.

Nevertheless I attempt review the paper with focus on other aspects. Papers’ structure and research are correct. The paper is complete. What I said the theoretical side is not clear for readers who are not very deeply involved into the topics. The parameters presented on graphs should have more clear explanation in straight relation to physical meaning. It will be profitable for generic readers.

Despite the simulation seem very promising the question is what about tests in reality?

 

The paper is carefully prepared

 

 

Concluding

Overall merit of the paper from my side is that it is worth of publishing.

Author Response

Dear Reviewer:

Thanks for the valuable comments. Please see the attachment.

Author Response File: Author Response.pdf