Multiple-Network-Based Control System Design for Unmanned Surveillance Applications

Round 1

Reviewer 1 Report

The paper is devoted to a system that can be used to control multiple agents, both stationary and mobile. Agents are controlled based on several communication methods: LoRa, Bluetooth, LTE and Wi-Fi forming a multi-network control system. However, I have some comments to authors, please find them below.
1) What is the advantage of using the proposed multi-network control system?

2) The use of 4 types of sensors increases both energy consumption and the weight of robotic agents, how justified is the use of such systems?

3) In the text of the article, there are no numerical estimates of the results obtained. To what extent is the proposed system better than similar ones? What are its benefits? Give approximate numerical estimates of what maximum areas such mobile agents can cover.

4) The conclusions need to be expanded, include numerical estimates, the significance of the obtained scientific results and describe everything more specifically.

5) Add a paragraph about contribution to the scientific field to the text of the article.

Minor issues:

1) Line 49." Although an interface with a monitoring robot system using wireless communication exists, its application to actual sites is limited [14]". Explain what are the limitations?

2) Figure 1. Text in blue boxes should be larger.

3) Table 1. What is LGU?

4) What is the link [16] for?

With all mentioned, I believe the work may be published after the major revision.

Author Response

The paper is devoted to a system that can be used to control multiple agents, both stationary and mobile. Agents are controlled based on several communication methods: LoRa, Bluetooth, LTE and Wi-Fi forming a multi-network control system. However, I have some comments to authors, please find them below.
1) What is the advantage of using the proposed multi-network control system?

Sol) Communication stability, which plays an important role in robot control, can be secured by ensuring communication diversity/polymorphism.

2) The use of 4 types of sensors increases both energy consumption and the weight of robotic agents, how justified is the use of such systems?

Sol) When communication is lost, there are situations in which recovery is possible with a simple remote reboot, so it is possible to respond using multiple sensors, and with a single sensor, it is rather a waste of energy and time because it is necessary to manually process and return to the mission. 

3) In the text of the article, there are no numerical estimates of the results obtained. To what extent is the proposed system better than similar ones? What are its benefits? Give approximate numerical estimates of what maximum areas such mobile agents can cover.

Sol) With the unmanned robot security system, it was possible to perform unmanned security for more than 48 hours in practice. Since the applied mobile agent has an operating time of 2 hours, security was performed by rotating multiple agents, and the coverage area of the mobile agent is a fan shape including 3D LiDAR 15m and RGB 90degree at a speed of 1m/s. However, mobile agents can apply coverage planning differently depending on the sensor system and speed.

4) The conclusions need to be expanded, include numerical estimates, the significance of the obtained scientific results and describe everything more specifically.

Sol) In the conclusion, I mentioned the thesis on expenses using the DB and probability map obtained through this study.

5) Add a paragraph about contribution to the scientific field to the text of the article. 

Sol) Add a paragraph

Minor issues:

1) Line 49." Although an interface with a monitoring robot system using wireless communication exists, its application to actual sites is limited [14]". Explain what are the limitations?

Sol) Add a explain

2) Figure 1. Text in blue boxes should be larger.

Sol) Apply a comment

3) Table 1. What is LGU?

Sol) revise 100km

4) What is the link [16] for? 

Sol) You can mention the multimodal sensor module and refer to the agent's sensor coverage.

With all mentioned, I believe the work may be published after the major revision.

Author Response File: Author Response.pdf

Reviewer 2 Report

Overall the work is decent contribution however below are some of my comments,

 

1. Please address the typos and the grammatical errors in this manuscript.
2. How is the performance of the proposed determined here? Have we introduced approach obtained outstanding performance with the measures of accuracy, sensitivity, and specificity and if yes, what are they
3. Which database was the data stored for analyzing and then coming to a conclusion in this manuscript ?

MySQL or something else ? What techniques/tools if any were used for ETL process if any ?

4. Please be advised that even small savings in response time can save many hours of  delay and we should highlight these details in this manuscript supporting with the data.
5. Can we please add "Future Directions" as a the final section so that we can address the below questions,
1. Future directions/recommendations to mitigate the risks ?
2. How much cost (monetary impact) will it happen after the issue has been resolved ?
6. Are these application designed with a system that has been carefully engineered to ensure safety in emergency situation ? If yes, can you please share the necessary details around it in this manuscript?
7. We should add some specific details around Privacy impacts and ethical issues raised by the technology in this manuscript.
8. Please be advised that some details pertaining to the quadrotor can be incorporated from the search and recognize some objects standpoint in addition to adding some details for location detection. Additionally, I do not see any details around Adaboost Classifier and Pinhole Algorithm in this manuscript. Can we please make sure to add them?
9. Can you please highlight the computed average error in the findings section for this manuscript ? I would also request you to add some supporting data to back it up.
10. For the efficient and reliable information sharing between many software components, a CORBA based software framework should be incorporated. Can you please make sure to include some details around it in this manuscript?
11. Additionally, please be advised a state transition should be triggered based on the  conditions of the sensor measurements
12. Persistent surveillance or monitoring does not have any mention in this manuscript and should be thoroughly looked at. Persistent surveillance can be thought of as a generalization of the patrolling problem and Patrolling is basically defined as surveillance around an area to protect or supervise it which often uses concepts such as idleness, refresh time, blanket time, or latency. Can we please build on this thought and add some details around it in this manuscript?
13. An inevitable part of persistent surveillance is the ability to support persistent operations in real life—in presence of uncertainties and failures. We should definitely add some details around it in this manuscript with some supporting data.
14. Have we identified any Results concern in this manuscript? We should add some information around the performances of each local image processing system along with the performances of the CDMA multimedia transmission system. Additionally, the experimental results show that the deep learning-based method performs best in terms of accuracy and stability.
15. Computer simulations should be conducted to demonstrate the performance of the proposed method along with providing some back up data to support the claims.
16. We should add a new subsection 'Remote Sensing Systems' where there are two primary types of remote sensing systems, active and passive remote sensing systems. In active remote sensing system, the sensors are responsible for providing the source of energy required to detect the objects. The sensor transmits radiation toward the object to be investigated, then the sensor detects and measures the radiation that is reflected from the object.

Author Response

Thanks for your comments

 

1. Please address the typos and the grammatical errors in this manuscript.

Sol) Native speaker proofreading was requested.

 

1. How is the performance of the proposed determined here? Have we introduced approach obtained outstanding performance with the measures of accuracy, sensitivity, and specificity and if yes, what are they

Sol) The communication performance was determined as a product with normal performance, and it was designed by distributing roles from the robot's operating point of view.

 

1. Which database was the data stored for analyzing and then coming to a conclusion in this manuscript ?

Sol) Referred to and utilized in references 21 and 22

 

MySQL or something else ? What techniques/tools if any were used for ETL process if any ?

Sol) We only used MySQL.

 

1. Please be advised that even small savings in response time can save many hours of  delay and we should highlight these details in this manuscript supporting with the data.

Sol) The concept was established as saving the most time by reducing the situation in which a person goes and the situation in which a robot returns unconditionally by overcoming the situation in which the robot cannot operate due to communication disconnection(Fig 14. 48 hours continuous operating).

 

1. Can we please add "Future Directions" as a the final section so that we can address the below questions,

1. Future directions/recommendations to mitigate the risks ?

  Sol) Modified as conclusion and future works. And, mentioned Iridium communication and wifi performance improvements.

1. How much cost (monetary impact) will it happen after the issue has been resolved ?

  Sol) Time costs are reduced and continuous expenses are possible depending on the mobile agent.

 

1. Are these application designed with a system that has been carefully engineered to ensure safety in emergency situation ? If yes, can you please share the necessary details around it in this manuscript?

Sol) The thesis mentions that it is designed to send a signal to LoRa communication for an emergency situation where communication is lost during monitoring.

 

1. We should add some specific details around Privacy impacts and ethical issues raised by the technology in this manuscript.

Sol) All data used in the study were conducted with individual consent.(Include public data)

 

1. Please be advised that some details pertaining to the quadrotor can be incorporated from the search and recognize some objects standpoint in addition to adding some details for location detection. Additionally, I do not see any details around Adaboost Classifier and Pinhole Algorithm in this manuscript. Can we please make sure to add them?

Sol) It is difficult to add because the content is mentioned in another paper.

 

1. Can you please highlight the computed average error in the findings section for this manuscript ? I would also request you to add some supporting data to back it up.

Sol) Continuous LTE data is sensor data such as images, and unfortunately there is no error analysis data. However, a database using the data has been opened in Ref. 21. The wifi network used as a robot command operated without errors in operation for 48 hours.

 

1. For the efficient and reliable information sharing between many software components, a CORBA based software framework should be incorporated. Can you please make sure to include some details around it in this manuscript?

Sol) Cloud control software has been developed for integration, but it is difficult to include it due to the achievements of other institutions.

 

 

1. Additionally, please be advised a state transition should be triggered based on the  conditions of the sensor measurements

Sol) Triggered automatically according to communication purpose.

 

1. Persistent surveillance or monitoring does not have any mention in this manuscript and should be thoroughly looked at. Persistent surveillance can be thought of as a generalization of the patrolling problem and Patrolling is basically defined as surveillance around an area to protect or supervise it which often uses concepts such as idleness, refresh time, blanket time, or latency. Can we please build on this thought and add some details around it in this manuscript?

Sol) Added 48-hour continuous surveillance application results.

 

1. An inevitable part of persistent surveillance is the ability to support persistent operations in real life—in presence of uncertainties and failures. We should definitely add some details around it in this manuscript with some supporting data.

Sol) Stacking data to reduce errors is mentioned in Reference 22 using Human In Loop.

 

1. Have we identified any Results concern in this manuscript? We should add some information around the performances of each local image processing system along with the performances of the CDMA multimedia transmission system. Additionally, the experimental results show that the deep learning-based method performs best in terms of accuracy and stability.

Sol) We have confirmed that there is no problem in identifying abnormal situations, including data for more than 6 months.

 

1. Computer simulations should be conducted to demonstrate the performance of the proposed method along with providing some back up data to support the claims.

Sol) mentioned in Reference 22 using Human In Loop

 

1. We should add a new subsection 'Remote Sensing Systems' where there are two primary types of remote sensing systems, active and passive remote sensing systems. In active remote sensing system, the sensors are responsible for providing the source of energy required to detect the objects. The sensor transmits radiation toward the object to be investigated, then the sensor detects and measures the radiation that is reflected from the object.

Sol) mentioned in Reference 22 using Human In Loop

 

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors present a system that can be used to control multiple agents based on multiple communication methods. The work is interesting and can be used for various applications, though there is a scope for detailed investigations on the method. Hence, the authors are suggested to include more detailed analysis with a improvement of the quality of the presented figures. Also, the authors are suggested to improve the presentation of the work as well as to include more explanation of the method before accepted for publication. 

Author Response

Thanks your comments.

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors propose an interesting surveillance system, based on networks, rovers, and cameras, for monitoring areas.

The manuscript is written in a well-form, it presents the research in the correct structure and it also contains numerous images, screenshots, and code sections that make this work quite complete and interesting.

However, there is a bit of a lack of scientific soundness. Starting from the Bibliography, there is a low number of cited works, and the majority of the references are even not so recent (e.g., from 2020 to 2022). The work seems to aim at only developing an application. For this reason, we suggest the authors to include, especially in the first section of the manuscript, some scientific subjects related to the proposal, such as the following:

Machine or Deep Learning usage ( Mishra, Balmukund, et al. "Drone-surveillance for search and rescue in natural disaster." Computer Communications 156 (2020): 1-10. )

Security on the data ( Avola, Danilo, et al. "Visual cryptography for detecting hidden targets by small-scale robots." International Conference on Pattern Recognition Applications and Methods. Springer, Cham, 2019. )

Security on the transmission ( Wang, Hui-Ming, et al. "Secrecy and covert communications against UAV surveillance via multi-hop networks." IEEE Transactions on Communications 68.1 (2019): 389-401. )

Ground and aerial vehicles comparisons ( Liu, Yao, et al. "Cooperative routing problem for ground vehicle and unmanned aerial vehicle: The application on intelligence, surveillance, and reconnaissance missions." IEEE Access 7 (2019): 63504-63518. )

Those are just some possible examples that could improve the scientific soundness of the manuscript and that could provide a wider discussion range about the contexts in which the proposed application is located.

No other issues have been detected.

Author Response

Thanks for your comments. I applied them all. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All comments were taken into account, minor errors were corrected, necessary explanations were added. Article is ready for publication.

Reviewer 2 Report

Thank you for addressing all my comments.

Dr. Shahani