Control Allocation Strategy Based on Min–Max Optimization and Simple Neural Network
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsNice and interesting manuscript. But the main idea is presented in a somewhat complex way.
The detailed comments are as follows:
• What is the main question addressed by the research? In that paper proposes a novel control allocation strategy tailored for servo-free, passive-hinge tilt-rotor octocopters. The presented method integrates min-max optimization with the force decomposition algorithm, handling actuator saturation while maintaining low computational complexity.• Do you consider the topic original or relevant to the field? Does it
address a specific gap in the field? Please also explain why this is/ is not
the case. In my mind it was a proposed and resolved original task. It addresses control allocation strategy tailored for servo-free, passive-hinge tilt-rotor octocopters. That task contains complex and interesting engineering solutions.
• What does it add to the subject area compared with other published
material? From numerous publications it is possible to admit that conventional multi-rotor unmanned aerial vehicles generate unidirectional thrust, inherently coupling their position and attitude control, thereby limiting maneuverability and efficiency in complex operational scenarios. In the manuscript, researchers study tilt-rotor UAVs employing vectored thrust achieve decoupled position and attitude control, significantly enhancing maneuverability and operational capabilities.
• What specific improvements should the authors consider regarding the
methodology? I think authors parallel with simple neural and min-max optimisation, may estimate other automatic controls principles. Such as non-linear gradient and hierarchical methods using very popular now AI techniques.
• Are the conclusions consistent with the evidence and arguments presented
and do they address the main question posed? Please also explain why this
is/is not the case. In the manuscript the proposed method integrates min-max optimization with a null space-based approach, leveraging the computational efficiency of FD and the saturation-handling capability of min-max optimization.
The simulation results highlighted the balanced thrust distribution achieved by both the min-max method and the neural network-fitted min-max method.
In general, the proposed control allocation strategy combines high computational efficiency, robust handling of actuator constraints, and reliable real-time performance.
• Are the references appropriate? Yes, the references are appropriate to providing literature analysis and high-quality research.
• Any additional comments on the tables and figures. Tables are absent in the proposed manuscript. Only 3 figures exist in the research, but all of them logically describe and represent engineering tasks and solutions. Comments on the Quality of English Language
In general, Ñ–sometimes the text is quite difficult to understand. It is advisable to show the manuscript to a researcher for whom English is a native language.
Author Response
Dear Reviewer,
I hope this message finds you well.
Please find attached the PDF document with our point-by-point responses to the reviewer comments.
Thank you for your time and consideration.
Best regards,
Kaixin Li
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors conduct research on the control allocation strategy based on a servo-free, passive-hinged tilt-rotor octocopter unmanned aerial vehicle (UAV). Both the research object and the research content show certain innovation. It can effectively deal with the problem of actuator saturation while maintaining low computational complexity. At the same time, taking into account the constraints of practical engineering applications, the proposed control method can operate efficiently on the flight control board. However, some questions still need to be answered:
- The control algorithm is specifically designed for this innovative object. Does the control algorithm still possess a certain degree of universality? When a new control object is replaced, is there still a possibility of application?
- Conventional tilt-rotor UAVs rely on tilt mechanisms, but they usually have wings, and flight can be achieved with as few as 2 motor-propeller systems. For the research object in this paper, there are no wings like those of a fixed-wing aircraft, and it seems to be improved from a rotor UAV. However, in order to achieve tilt flight without a servo mechanism, its power system has been changed from 4 sets to 8 sets. Can this really reduce the system weight? At the same time, with the power system becoming more complex, will the system robustness really be improved?
- It is recommended to carefully revise Section 5. The simulation results are piled up on a large graph, and there is a serious separation between the analysis part and the result display part. Moreover, in the simulation result part, only the phenomena corresponding to the simulation results are described. There is no analysis of why the saturation problem occurs and how the proposed algorithm solves this problem, and there is no connection back to the previous algorithm design part.
- Which simulation result supports the conclusion of efficient operation? Is there a comparison of the algorithm operation time? I haven't found it.
- The Min-max algorithm itself is not new, and it has relatively large conservativeness. How can this problem be solved?
6.It is recommended to modify the line types of the simulation diagrams. Taking Figure 3 ij as an example, the t1 curve is basically invisible.
7.The simulation video link provided in Section 5 does not correspond to the research object of this paper.
8.The publication years of quite a few references are around 2015 or even earlier, which means they date back quite a long time. Please cite the latest and representative references. For example, “Hierarchical Robust Model Prediction Control for a Long-Endurance Unmanned Aerial Vehicle”, et al.
Author Response
Dear Reviewer,
I hope this message finds you well.
Please find attached the PDF document with our point-by-point responses to the reviewer comments.
Thank you for your time and consideration.
Best regards,
Kaixin Li
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors carefully answered the reviewer's questions and made serious revisions. But there is still one issue that is difficult to reconcile.
(1) Regarding the min max algorithm, the author also acknowledged its disadvantages in the reply letter and compensated for these disadvantages through various means, achieving acceptable results. However, as an academic paper rather than an engineering report, why persist in non cutting-edge algorithms?
Author Response
Dear Reviewer,
Thank you for forwarding the reviewers’ comments. We have revised the manuscript accordingly and prepared a detailed point-by-point response.
Please find the revised manuscript and response letter in the attachment.
Best regards,
Kaixin Li
Author Response File: Author Response.pdf