A Cascaded and Adaptive Visual Predictive Control Approach for Real-Time Dynamic Visual Servoing
Round 1
Reviewer 1 Report
The article deals with the scientific problem already quite well systematized in the world literature and UAV community. However, I see the authors' great effort to propose an optimal approach to the problem. The article is quite well prepared. I will refer to the minor comments that should be made:
- line 104: “its” instead of “it’s”,
- line 171: parametrizing,
- figure 2: Quanser,
- table 1: Quanser,
- line 273: known,
- line 287: MATLAB,
- line 332: Simulink,
- line 336: Experimental,
- figure 9: Experimental.
Author Response
Point 1: The article deals with the scientific problem already quite well systematized in the world literature and UAV community. However, I see the authors' great effort to propose an optimal approach to the problem. The article is quite well prepared. I will refer to the minor comments that should be made:
- line 104: “its” instead of “it’s”,
- line 171: parametrizing,
- figure 2: Quanser,
- table 1: Quanser,
- line 273: known,
- line 287: MATLAB,
- line 332: Simulink,
- line 336: Experimental,
- figure 9: Experimental.
Response 1: We would like to thank the reviewer for their valuable comments. We have corrected the errors per reviewer’s. We also double-checked on typos and spelling errors. Please find the highlighted revisions in the PDF attached.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Reviewer’s comments on manuscript titled “A Cascaded and Adaptive Visual Predictive Control Approach for Real-time Dynamic Visual-Servoing”.
GENERAL COMMENTS
The authors have presented an interesting study outlining an approach for real-time dynamic visual-servoing. While such approach is not new, the study is still relevant for the scientific community and is acceptable for publication in the journal. The manuscript is written well, however, I have a few concerns which need to be addressed to make this manuscript acceptable for publication. For more detailed comments, please see the attached file.
Comments for author File: Comments.pdf
Author Response
The authors have presented an interesting study outlining an approach for real-time dynamic visual-servoing. While such approach is not new, the study is still relevant for the scientific community and is acceptable for publication in the journal. The manuscript is written well, however, I have a few concerns which need to be addressed to make this manuscript acceptable for publication. For more detailed comments, please see the attached file.
The authors would like to appreciate your comments and valuable feedback. We tried to improve on the paper per your comments. Please find the applied highlighted revisions in the PDF.
Point 1: Write a small section on the results pertinent to the study.
Response 1: The following section has been added to the abstract.
Simulated and experimental results show that the proposed method allows the model helicopter to servo toward the target efficiently in real-time while taking kinematic and dynamic constraints into account. The simulation and experimental results are in good agreement and promising.
Point 2: Is this a correct word?
Response 2: The correct word was “aggravate” which is replaced with the word “aggregate” in the revised manuscript.
Point 3: Start with Unmanned Aerial Vehicles and then use UAVs throughout the text.
Response 3: This is now implemented in the revised paper.
Point 4: Use a better word. utilized?
Response 4: We replaced the word “exploited ” with “employed”.
Point 5: To be consistent, use either UAVs or drones.
Response 5: In the revised paper, we use the term “UAVs” consistently throughout the paper.
Point 6: Is it a reference or equation? If equation, please use eq. 10.
Response 6: In the revised paper, we have added “Equation” to mention a reference to an equation based on your comment and that by the MDPI guidelines.
Point 7: Provide the details of the source.
Response 7: The following has been added to the manuscript in the revised version.
A 3-megapixel (Logitech C270) webcam [28] with a Complementary Metal-Oxide Semiconductor (CMOS) sensor,
Point 8: What will be the challenges in a real world scenario?
Response 8: We have pointed out the challenges associated with the target detection in un-structured real-world scenrios in the revised manuscript as follows:
In unstructured and real-world scenarios subjected to a cluttered background, the target in the image should be detected using more robust techniques such as those based on deep learning-based techniques. In this paper, the main focus was on the design of the controller, therefore, a simple blob-centroid-detection algorithm was used for tracking the target.
Point 9: Provide the details.
Response 9: We added some more details on Simulink and MATLAB implementation of the controller in the revised manuscript. Please see below.
The proposed control approach and image processing algorithm are implemented in Simulink®. The camera feed is interfaced to the Simulink model through Webcam Support extension package [35]. The image processing and feature detection algorithms have been developed using the functions and libraries provided in the Computer Vision Toolbox™ [ 36 ]. Furthermore, MATLAB functions and scripts were developed for the implementation of prediction models, control algorithms, and the Quadratic Programming (QP) optimization routine. The MATLAB functions were then converted into Simulink blocks through the process described in [37] for system integration.
Point 10: What about an irregular shaped object?
Response 10: Thanks for pointing this out. The problem of depth estimation for irregular shapes with unknown geometry could be challenging. We provide some references in the revised manuscript in this area. Some methods of choice would fall into: markerless object tracking/servoing, which was beyond the scope of this project.
the binary image was used to determine the coordinates of the centroid. Methods for estimating the depth of target for VS to irregular shapes can be challenging and have been extensively studied in [33,34].
Point 11: How will this algorithm perform in outdoor conditions with a lot of ambient noise?
Response 11: Thanks for the question. In fact, we are working on implementing the proposed control method on a free-flying multi-copter where our main challenge will be improving the control performance against exogenous disturbances such as wind gusts and providing robust control performance against noises and delays.
The spelling errors mentioned by the reviewer have also been corrected in the revised manuscript.
Please find the revisions in the highlighted PDF file attached.
The authors would like to thank the reviewer for their diligence and positive feedback.
Author Response File: Author Response.pdf
Reviewer 3 Report
The work proposed by the authors is novel and promising. They present with high efficiency their findings while the article's structure is excellent. Some minor language errors exist; however, I think that they will be corrected if the manuscript is accepted for publication.
Minor comments:
line 162: Dk is not shown in the equation.
line 187: the vectors u --> vector
line 190: represents --> represent
line 265: efficient thrust only rotating in one direction --> why?
line 279: sensor
line 331: are --> is
line 345: the experimental
line 346: in Figure
line 347: is proposed
line 359: "." The result
Author Response
Point 1: The work proposed by the authors is novel and promising. They present with high efficiency their findings while the article's structure is excellent. Some minor language errors exist; however, I think that they will be corrected if the manuscript is accepted for publication.
Minor comments:
line 162: Dk is not shown in the equation.
line 187: the vectors u --> vector
line 190: represents --> represent
line 265: efficient thrust only rotating in one direction --> why?
line 279: sensor
line 331: are --> is
line 345: the experimental
line 346: in Figure
line 347: is proposed
line 359: "." The result
Response 1: Thank you for your valuable feedback. We have corrected the above-mentioned errors. Please find the highlighted revisions in the PDF file attached.
line 265: efficient thrust only rotating in one direction --> why?
The following paragraph has been added to the manuscript to clarify.
The blades used in our model helicopter were designed by the manufacturer in a way that they could generate thrust efficiently only when rotating in a certain direction. In general, the propellers in rotary-wing flyers are designed to give only upward lift. For downward force, usually, it is left to the gravity. Our model helicopter was not an exception to this. The blades could provide lift efficiently only when turning in a certain direction. The aforementioned blade design would make our model helicopter more energy efficient. Downward motion was achieved by using the passive forces due to gravity, after the motors were switched off.
Author Response File: Author Response.pdf
