A Real-Time Simulator for Navigation in GNSS-Denied Environments of UAV Swarms
Round 1
Reviewer 1 Report
General - This article contains very timely information and is publishable after a lot more work.
The sentence structure is not good in many places but I am not supposed to correct English.
Once the system is described a significant amount of information is not referenced and it needs to be
as the authors did not develop for example the shown equations. The lack of references of this type
makes this not publishable in current form along with poor sentence structure.
In general I do not see any mention of a combined GNSS-IMU solution in a Kalman filter sense?
(1) Why is IMU not mentioned in the introduction?
(2) define "high precision" .01, .1, 1 m or otherwise define differently and define how one
guarantees "high precision"
(3) define spatiotemporal benchmarks as a very non standard term
(4) Why in intro are there only 2 references for "Simulation technologies"?
(5) Line 54 why is differential GNSS not used to improve 4.6 meters?
(6) Line 76-77 define quality of inertial, altitude, and image measuring devices
(7) Line 85 - you say SURF without defining what it means check this for all abbreviations
(8) Line 87-88 an example of a sentence that is not a sentence - I am not supposed to comment
on English but this happens in many places
(9) line 90 "lager"?? I think larger
(10) Line 92 - define how much improvement?
(11) Equation should be numbered examples are line 111-113
(12) Line 115 are you really used 258? You would lose quality due to lack of significant figures.
(13) You have a lot of equations that you did not derive, they need references.
(14) line 239-240 "unreliable in" needs example and/or references
(15) line 247 what does bsection4 mean?
(16) You never defined what is Gazebo?
(17) Figure 7 text covers the image.
(18) line 291 "about 0.75 m." - define about what is the randomness of 0.75 m. term
See previous comments there are lots of sentences that are not sentences. It needs a lot of English help.
Author Response
Dear professor #1
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements, as follows:
Reviewer 1.1:In general I do not see any mention of a combined GNSS-IMU solution in a Kalman filter sense?
answer: The main research content of the article is navigation in GNSS-denied environments.
Reviewer 1.2:Why is IMU not mentioned in the introduction?
answer: In line 61-66Added the following introduction about IMU.
“Due to factors such as changes in flight altitude and attitude during the flight process, the real-time aerial photos taken by drones exhibit complex geometric distortions , resulting in severe rotation, scaling, and even deformation compared to the pre stored reference images of drones. But the inertial information from inertial measurement unit (IMU) is not affected by external factors. So inertial information can be used to solve the perspective transformation relationship between images[13]. “
Reviewer 1.3:define "high precision" .01, .1, 1 m or otherwise define differently and define how one
guarantees "high precision"
answer: In scene matching navigation under GNSS-denied environments, navigation accuracy is related to flight altitude. Generally, a navigation error to flight altitude ratio of less than 1% is considered high precision.
Reviewer 1.4:define spatiotemporal benchmarks as a very non standard term.
answer: ”spatiotemporal benchmarks” means that accurate spatial position and velocity measurement relative to geographic coordinate systems.
Reviewer 1.5:Why in intro are there only 2 references for "Simulation technologies"?
answer: Added two references about Rflysim in line 49. Currently, 5 references on simulation technology are listed.
Reviewer 1.6:Line 54 why is differential GNSS not used to improve 4.6 meters?
answer: The system work in GNSS-denied environments, differential GNSS is not available.
Reviewer 1.7:Line 76-77 define quality of inertial, altitude, and image measuring devices
answer: This is the algorithm structure diagram, which only describes which data sources include and does not define sensor types.
Reviewer 1.8:Line 85 - you say SURF without defining what it means check this for all abbreviations
answer: Added “speeded up robust features (SURF)” in line 118.
Reviewer 1.9:Line 87-88 an example of a sentence that is not a sentence - I am not supposed to comment on English but this happens in many places
answer: The Sentence has improved as follow.” Inertial navigation system(INS) is used to reduce the number of reference images to improve the performance of SMN.”
Reviewer 1.10: line 90 "lager"?? I think larger
answer: This error has been changed.
Reviewer 1.11:Line 92 - define how much improvement?
answer: It is explained in the simulation results in section 4.1.
Reviewer 1.12:Equation should be numbered examples are line 111-113
answer: Equations in line 111-113 have been numbered.
Reviewer 1.13:Line 115 are you really used 258? You would lose quality due to lack of significant figures.
answer: The oblateness of the Earth is one of the main parameters describing its shape. According to the resolution of the International Association of Geodesy and Geophysics in 1971, .
Reviewer 1.14:You have a lot of equations that you did not derive, they need references.
answer: Some references related to the research of the article have been added.
Reviewer 1.15:line 239-240 "unreliable in" needs example and/or references
answer: The cited literature has been added at this location.
Reviewer 1.16:line 247 what does bsection4 mean?
answer: The correct expression is “section 4”.
Reviewer 1.17:You never defined what is Gazebo?
answer: Gazebo is a free robot simulation software that provides high fidelity physical simulations, a complete set of sensor models, and a very user-friendly interaction between users and programs.
Reviewer 1.18:Figure 7 text covers the image.
answer: This error has been corrected.
Reviewer 1.19: line 291 "about 0.75 m." - define about what is the randomness of 0.75 m. term
answer: The maximum error is 0.75m.
Yours sincerely,
Author
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The researchers in this paper have focused on the development and experimentation of a real-time simulator for navigation in GNSS-denied environments. There are many issues with the current version of this research paper as mentioned below. The following suggestions are made to improve this paper.
1. Although the authors have pointed out the advantages of using simulation technology in academic research and automatic flight control systems also in Introduction, several major concerns are still there as mentioned below.
a. Uses and importance of UAV formation, navigation as well as UAV swarm have not been stated properly.
b. Various issues with the ongoing researches on UAV formation and navigation have not been pointed out.
c. Authors’ contributions to address such issues have not been stated properly.
2. The abbreviated terms like ‘UAV’, ‘GNSS’, ‘SMN’ etc., must be expanded on its first use in the Introduction of the paper.
3. The word ‘GNSS’ mentioned at line 27 on page 1 should be replaced with ‘(GNSS)’.
4. Other researchers’ works must be cited in proper way. Instead of using bare reference no., i.e., 9, 10, 11 at lines 51, 54 and 57, [9], [10], [11] must be used to cite those references.
5. ‘Fig1’ mentioned at line 69 must be replaced by ‘Fig. 1’.
6. The word phrases ‘simulator contains of’ mentioned at line 69 must be replaced by ‘simulator contains’
7. The background and related work on UAV positioning & navigation in GNSS-denied environments must be provided.
8. The uses of various components such as formation control, controller, SMN, sensor models, relative navigation module, etc. shown in fig. 1 and how these components work together to implement UAV navigation should be explained in Section 2 for better understanding of the proposed architecture.
9. The working methodology along with flow chart for scene mapping navigation should be provided in subsection 2.1.
1. The performance metrics used to evaluate the performances of the proposed system/algorithm must be defined at first in Section 4.
1. The X axis of the graph given in fig. 9(b) must be labelled with measurement unit.
Comments for author File: Comments.pdf
Moderate editing required.
Author Response
Dear professor #2
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements, as follows:
Reviewer 2.1:Uses and importance of UAV formation, navigation as well as UAV swarm have not been stated properly.
answer: The uses and importance of UAV formation, navigation as well as UAV swarms have been stated as follows:
Advances in science and technology in the recent decade have brought increased interest towards the development of autonomous UAV. Swarm is general in nature, such as fish schools, ants and honeybee swarms. Study of UAV swarm based on biologically inspired have become the inevitable trends of UAV development[1][2]. This makes autonomous UAV a very captivating technology for activities such as geological mapping, surveillance, resource gathering, and rescue missions, among others. Maintaining, changing, and reconstructing formation control is necessary for UAV swarms missions. One fundamental component of UAV swarms systems is a navigation system capable of providing navigation and relative positioning. Currently, most UAV formation controls are based on integrated navigation systems which include an inertial measurement unit (IMU) and a global navigation satellite system (GNSS) receiver. The IMU can accurately estimate the attitude and position of the UAV over a short period of time. But due to random drift errors of inertial devices, the pose error accumulates over time. GNSS such as the global positioning system (GPS) is more accurate over longer periods of time. Therefore, IMU and GNSS are usually combined through extended Kalman filters (EKF) or unscented Kalman filters (UKF) to achieve high-precision pose information over long time.
Reviewer 2.2:Various issues with the ongoing researches on UAV formation and navigation have not been pointed out.
answer: The issues with the ongoing researches on UAV formation and navigation have been pointed out in line 40-46.
However, in constantly evolving high-intensity adversarial environments, relying solely on GNSS to establish reliable spatiotemporal benchmarks has become difficult[3]. Because GNSS has shortcomings such as extremely susceptible to interference, non-line-of-sight (NLOS) reception and signals spoofing and son on[4]. Therefore, navigation methods that do not rely on communication under GNSS-denied environments have become increasingly important for UAV swarm applications.
Reviewer 2.3:Authors’ contributions to address such issues have not been stated properly.
answer: The contributions has been stated as follows:
In this paper, a real-time simulator for navigation in GNSS-denied environments was proposed. The demo video webpage link is as follows:
https://v.youku.com/v_show/id_XNTk5NDIwMzA3Ng==.html?spm=a2hcb.playlist.page.1&playMode=pugv
The simulator integrates the world environment, sensor physical models, dynamic models of fixed-wing drone and multi-rotors drone, controller of the drones, formation controller, inertial-aided SMN(ISMN) module, relative navigation based on UWB and vision module. The contributions of this study are as follows:
(1) A real-time simulator for navigation in GNSS-denied environments is developed in order to improve the iteration efficiency of navigation algorithms.
(2) A novel scene matching navigation algorithm we called ISMN is proposed. Based on the simulator, the ISMN algorithm is validated.
(3) A relative navigation method that does not rely on inter communication is proposed.
Reviewer 2.4:The abbreviated terms like ‘UAV’, ‘GNSS’, ‘SMN’ etc., must be expanded on its first use in the Introduction of the paper.
answer: The abbreviated terms has been expanded on its first use.
Reviewer 2.5:The word ‘GNSS’ mentioned at line 27 on page 1 should be replaced with ‘(GNSS)’
answer: This error has been corrected.
Reviewer 2.6:Other researchers’ works must be cited in proper way. Instead of using bare reference no., i.e., 9, 10, 11 at lines 51, 54 and 57, [9], [10], [11] must be used to cite those references.
answer: This error has been corrected.
Reviewer 2.7: ‘Fig1’ mentioned at line 69 must be replaced by ‘Fig. 1’.
answer: This error has been corrected.
Reviewer 2.8: The word phrases ‘simulator contains of’ mentioned at line 69 must be replaced by ‘simulator contains’
answer: This error has been corrected.
Reviewer 2.9: The background and related work on UAV positioning & navigation in GNSS-denied environments must be provided.
answer: The background and related work on UAV positioning & navigation in GNSS-denied environments have been provided and some relevant literature has been cited.
Visual navigation calculates the position and attitude of UAVs relative to ground features to achieve UAV positioning and navigation. Visual navigation has the ad-vantage of minimal environmental impact and high accuracy, and has broad applica-tion prospects in GNSS-denied environments[7]. Benefiting from the simple structure, low cost, and high-precision positioning, scene matching navigation[8]system possesses a key prospect in the global navigation satellite system denied environment for un-manned aerial vehicles.
In [9] a novel SMN is proposed, by introducing an optimized factor of the homography matrix, the projection errors are reduced. In the experiment, the UAV flies in 150m height and the speed is 12m/s. The results show that the time consumption for on matching is 0.6 s, the position error is 4.6m. Due to factors such as changes in flight altitude and attitude during the flight process, the real-time aerial photos taken by drones exhibit complex geometric distortions, resulting in severe rotation, scaling, and even deformation compared to the pre stored reference images of drones. But the inertial information from inertial measurement unit (IMU) is not affected by external factors. So inertial information can be used to solve the perspective transformation relationship between images[16].
The work in [10], which investigates onboard visual relative localization in GNSS-denied environments, by detecting and localizing a simple circular pattern composed of concentric black and white circles of known diameter. Another approach [11] presents a novel onboard relative localization method for leader-follower formations of multirotor UAVs. The method uses a new smart sensor called UVDAR, which developed a UVDAR system, which can be used to obtain both the relative position and orientation by retrieving image positions and frequency-based IDs for individual blinking ultraviolet markers from a modified camera. In the literature [12], an onboard relative localization framework for multirobot systems, which utilizes UWB for peer-to-peer localization was presented
Reviewer 2.10: The uses of various components such as formation control, controller, SMN, sensor models, relative navigation module, etc. shown in fig. 1 and how these components work together to implement UAV navigation should be explained in Section 2 for better understanding of the proposed architecture.
answer: For better understanding of the Fig. 1, the workflow of the simulator descripted as follows.
In the simulation, the flight environment is set through the World module, the UAV flies in the environment with sensors such as accelerometer, gyroscope, magnetometer, UWB and camera and so on. The ISMN module collects the information of accelerometer, gyroscope, magnetometer and camera, and calculates the absolute pose of UAV. The pose information will be sent to Controller module, which can stabilize the UAV through closed-loop control. The Relative navigation based on UWB and vision module collects the image from camera and the distance from UWB, and calculates the relative position of the UAVS. The relative position will be sent to Formation controller module, which can implemente swarms control.
Reviewer 2.11:The working methodology along with flow chart for scene mapping navigation should be provided in subsection 2.1.
answer: Figure 3 shows the working methodology along with flow chart for scene mapping navigation.
As shown in Fig 3, the position of the UAV at time k is . After a short time, the position of the UAV at time k+1 is . The distance the UAV flies forward is which can be obtained by INS propagation. The deviation caused by drone attitude can be obtained through parameters such as drone flight attitude and altitude. The specific derivation process is as follows:
Reviewer 2.12:The performance metrics used to evaluate the performances of the proposed system/algorithm must be defined at first in Section 4.
answer: The simulation system can provide true values include absolute navigation and relative navigation. Compare the output of the navigation algorithm with the truth value provided by the simulation system to demonstrate the effectiveness of the algorithm. In line 320-322, the following description has been added.
To demonstrate the effectiveness of the simulation environment and localization algorithm by comparing the output of ISMN with the truth values provided by the simulation environment.
Reviewer 2.13:The X axis of the graph given in fig. 9(b) must be labelled with measurement unit.
answer: The axis of Figure 9 have added as shown in Figure 10.
Yours sincerely,
Author
.
Author Response File: Author Response.pdf
Reviewer 3 Report
The manuscript is interesting. It deals about a new proposed simulator as the solution of the problem of using GNSS in the control of UAVs.
Although no relevant comments are raised about the methodology followed, in my opinion authors should consider the following remarks before considering the manuscript for publication:
- The title should specify, in some way, the term “UAV”, because the present study is not valid for all types of means of transport.
- Introduction section should be extended with more numerous and updated researches.
- Authors should comment why GNSS systems have been proved to be unreliable in multiple contexts, which is the reason/goal of the present manuscript.
- The wording of the text should be revised. Some sentences do not make sense (e.g. Lines 86-88; line 248, …)
- Caption of Figure 7 is not in line.
- The axis and legend of Figure 9 are not readable.
- The caption of Figure 9 does not follow the Journal’s template “(a)” (“b) …
- The quality of Figure 11 must be improved. The lines, legends and axis are not readable.
- References in the text are not included according to Journal’s template.
Author Response
Dear professor #3
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements, as follows:
Reviewer 3.1 The title should specify, in some way, the term “UAV”, because the present study is not valid for all types of means of transport.
answer: The title has been changed to “A Real-Time Simulator for Navigation in GNSS-denied Environments of UAV Swarms”
Reviewer 3.2: Introduction section should be extended with more numerous and updated researches.
Answer: Some references related to the research of the article have been added.
Reviewer 3.3:Authors should comment why GNSS systems have been proved to be unreliable in multiple contexts, which is the reason/goal of the present manuscript.
Answer: The shortcomings of GNSS is provided and the goal of the manuscript is stated in line30-44.
One fundamental component of UAV swarms systems is a navigation system capable of providing navigation and relative positioning. Currently, most UAV formation controls are based on integrated navigation systems which include an inertial measurement unit (IMU) and a global navigation satellite system (GNSS) receiver. The IMU can accurately estimate the attitude and position of the UAV over a short period of time. But due to random drift errors of inertial devices, the pose error accumulates over time. GNSS such as the global positioning system (GPS) is more accurate over longer periods of time. Therefore, IMU and GNSS are usually combined through extended Kalman filters (EKF) or unscented Kalman filters (UKF) to achieve high-precision pose information over long time. Generally UAV swarms can achieve high-precision relative positioning by sharing location through inter networking. However, in constantly evolving high-intensity adversarial environments, relying solely on GNSS to establish reliable spatiotemporal benchmarks has become difficult[3]. Because GNSS has shortcomings such as extremely susceptible to interference, non-line-of-sight (NLOS) reception and signals spoofing and son on[4].
Reviewer 3.4:The wording of the text should be revised. Some sentences do not make sense (e.g. Lines 86-88; line 248, …)
Answer: The sentences of the manuscript has been reorganized.
Reviewer 3.5:Caption of Figure 7 is not in line.
answer: This error has been corrected.
Reviewer 3.6:The axis and legend of Figure 9 are not readable.
answer: The axis of Figure 9 has added as shown in Figure 10.
Reviewer 3.7:The caption of Figure 9 does not follow the Journal’s template “(a)” (“b)
answer: Figure 9 has changed to Figure 9 and Figure 10.
Reviewer 3.8:The quality of Figure 11 must be improved. The lines, legends and axis are not readable.
answer: The quality of Figure 11 has improved.
Reviewer 3.9:References in the text are not included according to Journal’s template.
answer: The references have been revised.
Yours sincerely,
Author
Author Response File: Author Response.pdf
Reviewer 4 Report
The manuscript deals with the development and testing of a framework for online positioning (Real-time) in areas that do not have Global Positioning Systems (GNSS) coverage.
The comments for paper quality improvement are listed below:
General Comments:
1- In the abstract, the statement of the problem, research innovation (compared to previous studies), and the obtained results should be expressed better and more clearly.
2- The referencing is very unfavorable, and they mislead the reader.
3- Research innovation is explained very little in the Introduction. Also, the advantage of the proposed method compared to former studies should be clarified in the final part of the introduction section.
4- The analysis and results are not enough. To improve the article, more experiments numerical analysis, and graphs can be used.
5- The number of references used in the article is not sufficient.
Others:
6- It is mentioned in line 72 that Google Earth images are used as a database for scene matching. While Google Earth images have low spatial absolute accuracy, is it right to use them as a database to determine the absolute position?
7- In Figure 1, the fonts are inappropriate and disproportionate, correct them.
8- In Figure 2, some abbreviations such as IMU are used, which are not fully mentioned in the text before this figure.
9- In some parts of the text, writing modifications such as punctuation are required, for example, in lines 145 and 187, a dot is not placed at the end of the sentence.
10- Section 3.1 is started without a title for section 3
11- In line 219, the reference is wrongly embedded.
12- Figure 7 is very inappropriate and there is no need to have image margins.
13- The quality of Fig. 11 is poor; replacing it with an enhanced one is needed.
14- KF Filter is mentioned in line 171. This filter needs to be fully explained in the context.
15- The Hungarian algorithm is only mentioned in line 172, it needs to be briefly explained in the context.
In some parts of the text, writing modifications such as punctuation are required,
Author Response
Dear professor#4
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements, as follows:
Reviewer 4.1: In the abstract, the statement of the problem, research innovation (compared to previous studies), and the obtained results should be expressed better and more clearly.
Answer: The abstract of the article has been revised as follows.
Unmanned aerial vehicle (UAV) swarms require accurate localization to safeguard flight missions in the global navigation satellite system(GNSS) denied environment. The implementation of UAV formation in the real world is costly and time consuming. A real-time simulator for navigation in GNSS denied environments was proposed, which includes world, model, controller, scene matching navigation (SMN), relative navigation and formation controller module. Each module can be modified, which means that users can test its own algorithm. A novel inertial-aided SMN(ISMN) algorithm was developed and a relative navigation method that does not rely on inter communication was proposed. ISMN and relative navigation based on camera and Ultrawideband (UWB) were tested on the platform. By using the developed simulation system, the navigation algorithms can be verified easily and save substantial effort in flight tests.
Reviewer 4.2: The referencing is very unfavorable, and they mislead the reader.
Answer: Some references related to the research of the article have been added.
Reviewer 4.3: Research innovation is explained very little in the Introduction. Also, the advantage of the proposed method compared to former studies should be clarified in the final part of the introduction section.
Answer: The research innovation is explained in the introduction section.
In this paper, a real-time simulator for navigation in GNSS-denied environments was proposed. The demo video webpage link is as follows:
https://v.youku.com/v_show/id_XNTk5NDIwMzA3Ng==.html?spm=a2hcb.playlist.page.1&playMode=pugv
(1) A real-time simulator for navigation in GNSS-denied environments is developed in order to improve the iteration efficiency of navigation algorithms.
(2) A novel scene matching navigation algorithm we called ISMN is proposed. Based on the simulator, the ISMN algorithm is validated.
(3) A relative navigation method that does not rely on inter communication is proposed.
Reviewer 4.4: The analysis and results are not enough. To improve the article, more experiments numerical analysis, and graphs can be used.
Answer: In section 4, experiments numerical analysis is added.
Reviewer 4.5: The number of references used in the article is not sufficient.
Answer: The references have been revised.
Reviewer 4.6: It is mentioned in line 72 that Google Earth images are used as a database for scene matching. While Google Earth images have low spatial absolute accuracy, is it right to use them as a database to determine the absolute position?
Answer:
In scene matching navigation system, usually using satellite image as a benchmark map. Google high-definition satellite map is the most commonly used satellite map as a free map. The accuracy of maps is related to the level of division. Map tiles refer to a technique of dividing global map data into small tiles. Each tile has a unique identifier and location information, and is divided into multiple levels for display at different zoom levels. This slicing technology allows users to load and browse map data faster, while also reducing the burden on the server. Generally, tile maps define a zoom level within the range of 1-20 levels. The projection plane range covered by each zoom level remains unchanged, and tiles are segmented according to different display ratios and ground resolutions, resulting in a tile pyramid model with a decreasing number of tiles and a decreasing map size. The resolution of the 19 level Google tile map is 0.3m and the 18 level Google tile map is 0.5m. For GNSS denied environments, this resolution is sufficient.
Reviewer 4.7: In Figure 1, the fonts are inappropriate and disproportionate, correct them.
Answer: This error has been corrected as shown in Figure 1.
Figure 1. Architecture of the simulator
Reviewer 4.8: In Figure 2, some abbreviations such as IMU are used, which are not fully mentioned in the text before this figure.
Answer: In line 131-132, the role of IMU has been explained
Using IMU propagation to reduce the number of reference images to improve the real-time performance.
Reviewer 4.9: In some parts of the text, writing modifications such as punctuation are required, for example, in lines 145 and 187, a dot is not placed at the end of the sentence.
Answer: This error has been corrected.
Reviewer 4.10: Section 3.1 is started without a title for section 3.
Answer: The title for section 3 is “Configuration of Simulator”.
Reviewer 4.11: In line 219, the reference is wrongly embedded.
Answer: The references have been revised.
Reviewer 4.12: Figure 7 is very inappropriate and there is no need to have image margins.
Answer: This error has been corrected
Reviewer 4.13: The quality of Fig. 11 is poor; replacing it with an enhanced one is needed.
Answer: For better understanding, the Fig.11 has been adjusted.
Reviewer 4.14 KF Filter is mentioned in line 171. This filter needs to be fully explained in the context.
Answer: The state equation and observation equation of KF filter have been added.
For the KF filter of trajectory predicted, the state of the system is,
(16)
Where and is the center of the object in the image, is the aspect ratio, is the
altitude, the observation equation can be expressed as:
(17)
Reviewer 4.15: The Hungarian algorithm is only mentioned in line 172, it needs to be briefly explained in the context。
answer: The explain of the Hungarian algorithm has been added.
“And, using Hungarian algorithm which is a matching algorithms to match the detected and predicted result.”
Yours sincerely,
Author
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
1.1 The authors still need to discuss GNSS denied could potentially include some GNSS just not enough for a fixed solution.
1.3,1.4,1.7,1.17,1.19 - they answered me directly but my comments required editing of their article
1.13 by modern standards this is an outdated ellipsoid - why do they need to use an outdated ellipsoid
An English expert needs to review
Author Response
Dear professor #1
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements. The response to your review comments is specifically described in the attachment.
Yours sincerely,
All Authors
Author Response File: 
Author Response.docx
Reviewer 2 Report
Authors have uploaded a revised manuscript showing all track changes made on their initial manuscript using some MS word features based on the reviewers’ comments. It is very hard to follow such revised manuscript, specifically, it’s not at all clear how the authors have updated their original manuscript to address each concern/query raised by the reviewers.
Authors are suggested to provide a letter in which point-to-point reply to each concern/query of the reviewers along with the updated texts/parts of the revised manuscript must be presented. Moreover, authors should upload the revised manuscript with the highlighted changes along with the above-said letter.
Author Response
Dear professor,
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements. The response to your review comments is specifically described in the attachment.
Yours sincerely,
All Authors
Author Response File: Author Response.docx
Reviewer 4 Report
Accept
Author Response
Dear professor,
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements.
Yours sincerely,
All Authors
Round 3
Reviewer 2 Report
The authors, in this article, propose a real-time simulator that can be used for navigation purposes in the GNSS-denied environment and also a scene matching navigation algorithm based on the proposed simulator. However, the present version of the article has several issues based on which the following suggestions are made to further improve its contents.
1. The term ‘SMN’ must be expanded on its first use.
2. The word ‘implemente’ given at line 130 must be changed to ‘implement’
3. Most of the references (e.g., ref. 5-9, ref. 10, 11, 18, 19, 20) must be represented using normal text instead of superscript form when cited in the paper.
4. The working principle of the proposed scene mapping navigation algorithm is shown in figure 3. However, a separate work flow diagram for this proposed algorithm should be provided for its better understanding.
5. The mathematical formulation for determining absolute error, theta_err, phi_err and so on, which are used for performance evaluation of the proposed system, should be provided in Section 4.
6. More discussion on experimental results is required.
7. Resolution of figure 12 must be improved and the measurement unit of X axis label (i.e., time) in figure 12 must be mentioned.
Comments for author File: Comments.pdf
Author Response
Thank you for your valuable comments on this work. The paper has been made
necessary changes according to the requirements, as follows:
Reviewer 2.1: The term ‘SMN’ must be expanded on its first use..
answer: We agree with the Reviewer's comment. This error has been corrected.
In line 13:12-14
A real-time simulator for navigation in GNSS denied environments was proposed, which includes world, model, controller, scene matching navigation(SMN), relative navigation and formation controller module.
Reviewer 2.2: The word ‘implemente’ given at line 130 must be changed to ‘implement’.
answer: We agree with the Reviewer's comment. This error has been corrected.
Reviewer 2.3: Most of the references (e.g., ref. 5-9, ref. 10, 11, 18, 19, 20) must be represented using normal text instead of superscript form when cited in the paper.
answer: We agree with the Reviewer's comment. This error has been corrected.
Reviewer 2.4: The working principle of the proposed scene mapping navigation algorithm is shown in figure 3. However, a separate work flow diagram for this proposed algorithm should be provided for its better understanding.
answer: Figure 3 shows the “Reference map from INS propagation”，which is one of the steps scene matching navigation. The work flow diagram for this proposed algorithm as shown in Figure 2. The specific process description is as follows:
In line-140-146:
The ISMN procedure as shown in Fig. 2. Using IMU propagation to determine the scope of the reference satellite map to improve the real-time performance. Considering that the real-time images from camera on board and reference images from satellite map are not in the same frame, the larger of the oblique, the higher the probability of mismatching. To solve this problem, we use the attitude from the IMU to correct the real-time input images. This approach enhances the accuracy of image matching in large viewing angles and, consequently, improves the precision of SMN.
Reviewer 2.5: The mathematical formulation for determining absolute error, theta_err, phi_err and so on, which are used for performance evaluation of the proposed system, should be provided in Section 4.
answer: The simulation system can provide true values for all states, the results from camera and UWB compare with the true values can obtain the absolute error.
In line 371:
Using the truth values provided by the simulation system as a reference, according to Fig. 12, the maximum relative position error is 0.75m(<0.75% of the distance between adjacent drones).
Reviewer 2.6: More discussion on experimental results is required.
answer:
In line 371-377:
Using the truth values provided by the simulation system as a reference, according to Fig. 12, the maximum relative position error is 0.75m(<0.75% of the distance between adjacent drones). According to (19), the accuracy of the relative position mainly de-pends on the line of sight and azimuth angle measured by camera. Because the error from UWB is only 10cm which is a small amount compared to the distance between machines. On the other hand, under the condition of fixed angle error, the farther the distance between drones, the greater the relative navigation error.
In line 393-396:
In future research, we will tackle this challenge from the perspective of multirobot co-operative optimization in continuous time domain. Such as, the relative navigation is closely related to the formation structure. Designing formation based on relative nav-igation accuracy is worth in-depth research.
Reviewer 2.7: Resolution of figure 12 must be improved and the measurement unit of X axis label (i.e., time) in figure 12 must be mentioned.
answer: We agree with the Reviewer's comment. This error has been corrected.
Author Response File: Author Response.pdf
