Next Article in Journal
Large-Scale Earthwork Progress Digitalization Practices Using Series of 3D Models Generated from UAS Images
Previous Article in Journal
Accuracy Assessment of Cultural Heritage Models Extracting 3D Point Cloud Geometric Features with RPAS SfM-MVS and TLS Techniques
 
 
Technical Note
Peer-Review Record

Design and Implementation of Intelligent EOD System Based on Six-Rotor UAV

by Jiwei Fan, Ruitao Lu *, Xiaogang Yang, Fan Gao, Qingge Li and Jun Zeng
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Submission received: 9 November 2021 / Revised: 4 December 2021 / Accepted: 9 December 2021 / Published: 11 December 2021

Round 1

Reviewer 1 Report

Dear authors of Design and implementation of intelligent EOD system based on six-rotor UAV,

You have an interesting case study to detect and remove small duds or explosive devices from the firing range by using an UAV system which has a gripper that can be lowered with a motor controlled cable retractor. Thank you for the corrections you have already made for the initial version. However, there is still plenty of things to fix. Moreover, being a technical note, you should be able to present the proposed method such that the reader 1) can easily understand how you system works, 2) can verify that the proposed solution is functional and plausible, and 3) could reproduce similar outcome.

Most importantly, the work requires proper proofreading which includes also consideration for the used terminology. Some of the sentences are really difficult to understand.  Moreover, some technical terms such as steering gear, coaxial take-up turntable, take-up rotating shaft, cable take-up device, bomb dropping operation area, BottleNect, neck, "airborne sky end map transmission", should be reconsidered. For example, based on the red bounding box in Figure 5, I believe the term "steering gear" should be something like a "servomotor" or a "gear motor" (At least I know that the original term, the steering gear is something totally different). 
For the winch (you call it with a term "winding cable retracting device"), I propose you rewrite the whole section since now the co-axial things and "take-up turntable" are really confusing terms.

Then also Section 3.1.3 Dual-vision-integrated PTZ pod should be rewritten. The sensor most likely is TSHD10T3 (instead of SHD10T3 you state in the manuscript, see TSHD10T3 on page 29 on http://www.topotek.com/product_description-en/TOPOTEK_GIMBAL_EN_20200721.pdf ). Please include there the true image information for the color and thermal image frames you were able to stream online from the UAV. The product description states that the sensor broadcasts full-HD color image and can augment the thermal data into the color channels. However, you state in the text that sensor has "4M effective pixels", but you should explain more accurately what you are able to extract from the sensor.

Although you have rewritten Section 3.1. Hardware design, Figure 4 is still confusing. The reason for the blue arrows is not clear. If the reader considers it as a flow chart, it now seems that the loop starts from the dual mode pod, and in any route you select, you end up to the dud on bottom left corner. Please, redesign the figure to show the hardware structure. You could for example use a larger figure of the drone (e.g., Figure 13) and use text and lines on a figure to pinpoint different subsystems. From these subsystems you could draw arrows to indicate different data flow, but please use different colors for different type information (image stream with different color than the command or data link). Note that you should indicate communication to both directions if there is any communication backwards (e.g. telemetry in the data link, which you call with a term "digital communication"). Moreover, you could also have a picture of the ground station to show the computer, software, and data links that are used to control the UAV. This could be easily integrated to the same figure such that on the left side you would have the remote operation side and on right side you could have the UAV and mounted equipment and then arrows indicating communication going from left to right and from right to left (with included text to indicate purpose of the arrow). After this, the hardware figure would have similar overall structure than the system workflow in Figure 3.

Please also rewrite Sction 3.2.2 and Figure 10. the BottleNect and other terms in Figure 10 are not explained in the text and the neck and prediction terms in the text are not understandable in the current context. Furthermore, why is the feature pyramid drawn and opened but other parts of the YOLOv5 are not explained? Would it make more sense to explain the overall structure of the used network better and just site to the works if you are using the network without modifications? If you have done your own modifications to YOLOv5, then you should tell exactly what parts you have modified and how.

Moreover, please discuss the image resolution dropping. Now you are most likely reducing your capacity to detect small objects since you reduce the image size before you feed it into the YOLOv5. Why is that needed? Why don't you use the full resolution images you receive from the camera (e.g. fullHD)?

In addition, please also describe in the work the target flying altitude to detect the grenades. Discuss what is the maximum height you are able to detect a hand grenade reliably. Does the camera resolution or zoom setting have an effect on this?

Furthermore, in Figure 12 add axis labels. Is the x axis training rounds? Furthermore, the image labels: objectness, precision, recall val GloU, val Objectness, [email protected] and [email protected]:0.95 are not defined in the work.

In addition, the text section after Figure 13 is confusing. What do you mean with the ariborne sky end map transmission? Also the "ground receiving end" is not a good term. Please define better terms throughout the text.

Please rewrite the sentence: "The ground station detected and recongnition the real-time video to assist in locating the location of the dud, and the UAV flying hand control platform system completed the follow-up process, i.e., grasping and transfer."

As a final note, please also proofread your abstract. It seems to have an sentence that has been edited but not finished: "The experiments on the explosives defusing are performed, and the results show that our system." 

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Reviewer 2 Report

English is significantly better than last time. I have only one notice: 

Line 229: „Figure 6. Mechanical gripper“ – it should be ond same page with the picture.

Author Response

Dear the Associate Editor and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Design and implementation of intelligent inspection and alarm flight system for epidemic prevention” (Manuscript ID: drones-1478127). Those comments are all valuable and helpful for revising and improving our paper as well as for guiding our future research projects. The revised portions are marked in green in the revised manuscript. Our responses to the associate editor’s and reviewers’ comments are shown below.

 

--------------------------------------------------------------------------------------------------

Responses to the associate editor’s and reviewer’s comments:

--------------------------------------------------------------------------------------------------

 

Reviewer #2:

 

Comment 1: English is significantly better than last time. I have only one notice: Line 229: Figure 6. Mechanical gripper – it should be on same page with the picture.

Response: Thank you for your critical and good comments. We have put line 229: "Figure 6. Mechanical gripper "- on the same page as the picture.

The detailed revision can be found in the section “3.1.2 Design of mechanical gripper”, on Lines 228, Page 6.

 

Special thanks to you for your helpful comments.

--------------------------------------------------------------------------------------------------

Author Response File: Author Response.doc

Reviewer 3 Report

The research performed here has the potential to be applied in many other scenarios outside of just military UXO removal. It would be wonderful to see some other applications mentioned in the conclusion. There remain several areas where the wording could be fixed, along with perhaps some word choices, but I will leave that to you and the editors to resolve. (See attached PDF with comments)

Comments for author File: Comments.pdf

Author Response

Dear the Associate Editor and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Design and implementation of intelligent inspection and alarm flight system for epidemic prevention” (Manuscript ID: drones-1478127). Those comments are all valuable and helpful for revising and improving our paper as well as for guiding our future research projects. The revised portions are marked in green in the revised manuscript. Our responses to the associate editor’s and reviewers’ comments are shown below.

 

--------------------------------------------------------------------------------------------------

Responses to the associate editor’s and reviewer’s comments:

--------------------------------------------------------------------------------------------------

 

Reviewer #3:

 

Comment 1: The research performed here has the potential to be applied in many other scenarios outside of just military UXO removal. It would be wonderful to see some other applications mentioned in the conclusion. There remain several areas where the wording could be fixed, along with perhaps some word choices, but I will leave that to you and the editors to resolve.

Response: Thank you for the helpful advice. This paper designs an intelligent EOD based on a six-rotor UAV. The system is a UAV platform with comprehensive functions and high integration. Combined with UAV autonomous navigation, deep learning, and other technologies, intelligent explosion detection and EOD tasks can be completed safely and efficiently. The experimental results show that the system has good practicality and can ensure the powerful promotion of EOD to the greatest extent. In the future, the system can also be widely used in many aspects such as emergency early warning, battlefield reconnaissance, material sampling in dangerous areas, disaster relief, and express delivery. There are many grammatical and spelling mistakes in the original paper. We are sorry for the inconvenience. Based on the reviewer’s comments, we reviewed the entire manuscript and tried to eliminate all grammatical and spelling mistakes. The manuscript was also edited by English editors recommended by MDPI journal to improve their English. Major changes to the original are highlighted in green in the revised version.

The detailed revision can be found in the section “5. Conclusion”, on Lines 412-415, Page 13.

 

Special thanks to you for your helpful comments.

--------------------------------------------------------------------------------------------------

 

Author Response File: Author Response.doc

Round 2

Reviewer 1 Report

Dear authors of Design and implementation of intelligent EOD system based on six-rotor UAV,

Thank you for all the corrections you have made! The work is now significantly better than the two first revisions! Especially, thank you for proofreading the text and writing it in understandable English. I liked you new term UXO and wide usage of the term thorough the work. Other terminology seems also a lot better now. Moreover, the updated figures (4 and 10) are better now. Thank you also for adding all the missing details to the winch, PTZ pod, and for the YoloV5 network. Thank you for also adding the normal flying altitude of 3 meters in the work, now I understood that you are doing really close proximity remote sensing, and not anything related to normal drone remote surveying imaged from 20 or more meters high.

I hope the next time you could proofread your work already before sending it for review because the quality of the presentation improved so much during it.

 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.


Round 1

Reviewer 1 Report

Dear authors of Design and implementation of intelligent EOD system based 2 on six-rotor UAV,

You have an interesting case study to detect and remove small explosive devices from the shooting ground by an UAV system which has a gripper that can be lowered with a motor controlled cable retractor. However, although being a technical note, you should still be able to present the proposed method such that the reader 1) can easily understand how you system works, 2) can verify that the proposed solution is functional and plausible, and 3) could reproduce similar outcome.

Unfortunately, the work is now difficult to read and some proofreading is required. There are currently a lot of confusing terms which are not explained at all, such as, the sky end transmitter, airborne cable take-up device, coaxial take-up turntable, take-up rotating shaft, cable winding device, limit stud, etc. You also seem to use multiple terms for the same things since sometimes you talk about duds or dangerous goods and sometimes you deal with grenades. Similarly sometimes you talk about cable winding device and sometimes cable retracting device. Please try to be as exact as possible so that the reader knows when you talk about the same thing to distinguish the situation when you are actually explaining an another device or part of the already explained device. 

Furthermore, some sentences are really difficult to understand. For example, see the last sentence of the first paragraph of Section 3: "Therefore, only the output interface from the visual image processing results meets the requirements of the input interface of the flight control system, It can easily import the relevant algorithms of visual image processing into the system while reducing the workload of the flight main controller, and there is no need to replace the software or hardware in the visual system, which can improve the real-time performance of the whole system." 

Also the figures are confusing. For example, in Figure 3 there are different shaped and colored boxes, but no explanation is given for the different colors or shapes in the diagram. Currently, the figure induces more questions than it answers. Similarly, the flow chart in Figure 4 is confusing. The black text is difficult to read on blue arrows and the meaning and ordering of the arrows is confusing. Are the arrows indicating a data flow, usage order, causal order, or what? Please explain your figures in text and please also give some information already in figure captions.

In addition to the clarity of presentation, there are also other deficiencies in the manuscript. Since you are proposing an UAV for object search and detection, you should shortly review similar works done on the same and related fields. There is plenty of comparable research done on image based object detection on drones which is related on your work. This related work can be found for example on the fields of drone based search and rescue research (see e.g. a review by Grogan, et al, 2018), or low-altitude UAV object detection (see e.g. a survey by Mittal, et al, 2020), or even in precision forestry (e.g., Diez, et al. 2021). Furthermore, as you also have the manipulation aspect in your paper, you should also shortly review the available manipulators already proposed for small UAVs (see e.g. the review by Xilun, et al., 2019).

In addition, the contents selected for the paper are questionable. Why would you explain the software architecture and how easy it is to create new processes. In my opinion it is out of focus, since a UAV picking hand grenades on a field does not need these properties. Instead you should focus to the story line that describes what is needed in the proposed  intelligent explosive ordnance disposal UAV and why. Moreover, you should add some test and results which numerically measure the effectiveness of the proposed solution in comparison to the standard operation currently performed. This could be organized for example by organizing a test setup where the drone system is tested in a controlled test scenario against the manual work currently done. Used time, success rate, and risky situations could be counted in both cases to make numerical comparisons. Also accuracy of most important subsystems such as image based grenade detection could be measured on a test scenario (through detection accuracy or false positive rate). These tests would make the manuscript a lot more trustworthy and valuable for the reader.


References

Grogan, S., Pellerin, R., & Gamache, M. (2018). The use of unmanned aerial vehicles and drones in search and rescue operations–a survey. Proceedings of the PROLOG.

Mittal, P., Sharma, A., & Singh, R. (2020). Deep learning-based object detection in low-altitude UAV datasets: A survey. Image and Vision Computing, 104046.

Diez, Y., Kentsch, S., Fukuda, M., Caceres, M. L. L., Moritake, K., & Cabezas, M. (2021). Deep Learning in Forestry Using UAV-Acquired RGB Data: A Practical Review. Remote Sensing, 13(14), 2837.

Xilun, D. I. N. G., Pin, G. U. O., Kun, X. U., & Yushu, Y. U. (2019). A review of aerial manipulation of small-scale rotorcraft unmanned robotic systems. Chinese Journal of Aeronautics, 32(1), 200-214.

 

Reviewer 2 Report

The English text needs to be improved (e.g. humans beings to work need to be replaced with humans working). There is word document with corrected english for Abstract and 1st chapter. Please do it for the whole text.

Comments for author File: Comments.docx

Reviewer 3 Report

The research conducted for this paper demonstrates yet another ingenious form of applied UAS research that combines both UAS technology and AI. I applaud the authors for their research, and for publishing in a journal that is likely not the primary language. That stated, the quality and novelty of this research are difficult to grasp due to the research team needing somebody who can help make writing more understandable and better describe the methods. 

I am confident once this is done, the writing will convey the quality of the research performed here, and a much better review can be conducted.

I have attached a PDF that contains suggested comments throughout the document as a means to improve upon the writing throughout the manuscript.

 

Comments for author File: Comments.pdf

Back to TopTop