UAV Sensors Autonomous Integrity Monitoring—SAIM
Round 1
Reviewer 1 Report
- Why are the sensors selected to be 5? If just random number, you can make the equations general as a function in number of the sensor, then you applied the special case 5 as number of sensors.
- How the sensors distributed?
- You stated that the sensors are the same type and the decisions regarding the physical quantities, what is the reason behind the big variation for quality characteristics ( Pf, Pd) in table(1).
- I do know, but I feel that you need to add a brief description for SAIM before distinctive stages.
- Adding more details about fig.2, 3. The x- access indicating to threshold ( -5 to 5), is this representing the “threshold number” in table2,3?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Kindly provide some examples regarding experimental results, it was not very clear. Some more results to understand the application will be appreciated. On page 7, there is a minor typographical error where instead of "it" "if" is written. It is advisable to proofread the manuscript to avoid any typos.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper proposes the SAIM method for monitoring the operation integrity of each one of the sensors on UAV. In section 1, the background of collision avoidance and the sensor integrity are given. Next, the collision avoidance of UAV is introduced. In section 3, the author presents the RAIM essentials. The proposed SAIM is described in section 4 and finally the proposed methods are validated through experiment.
This paper is worth affirming, but it still has some problems:
1. This method has moderate innovation in mathematics.
2. The article has structural problem. It doesn’t clarify the purpose of writing the section 3, the RAIM essentials. It seems that the RAIM essentials are more like parts of the Introduction.
3. The experiments are not sufficient to fully validate the effectiveness of the proposed method for deployment on UAV. Only a set of data is assumed for experiment, and there is a lack of simulation or real experiments of the proposed algorithm on UAV scenarios.
a) The data of the experiment is simple and the assumptions are ideal (line 275 and 276).
b) It is recommended to build the testing scenarios in the simulation software to test SAIM. The reviewer would prefer to see the effect of the proposed method in a real collision avoidance situation.
c) No baseline is selected and compared with the proposed SAIM to highlight its advantage.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The manuscript looks much better.
