Research on Cooperative Stabilization Control of Multi-Pointing-Mirror Laser Communication Terminals Based on GA-ADRC
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsIn the paper, the authors use a correct English found in typical research papers. Focus on methodology and algorithm description sections, where some overlongish sentences could be splitted into other. Try to rewrite also the sentence with 'high precision stabilization accuracy' - can accuracy be of low precision?
The ADRC technique is well-established, thus it is natural that some issues can be found in the extisting literature. Tjhe combination of GA with ADRC for multi-point mirror stabilization is a novel application of known techniques. Some parts of the algorithm originate directly from GA or ADRC methodologies, however, everything is properly cited. Please ensure proper cross-referencing to your prior papers is clearly visible.
In the paper, the authors aim at overcoming challenges related to nonlinearity, time-varying parameters, and multi-channel coupling in a stabilisation task. In the introductory part of the paper, the area of interest is well defined, and compared to recent advancementts in the field. From this point of view, PID approaches are regarded as inadequate, thus the ADRC seems to be the only solution. in order to optimise ARDC parameters, GA is used to operate under complex disturbances.
The main novelty of the paper is a novel integration of GA-optimized ADRC in a distributed control architecture for multi-LOS stabilizationm iffering adaptation capabilities and simultaneous disturbance rejection.
A multiobjectice approach is used to define the fitness function for GA that takes stabilization accuracy, speed, and disturbance rejection into account. What is more, the authors conduct both simulations, as well as experimental tests to present their approach.
In the theoretical layer, the ERSO, TD, and nonlinear feedback are coupled in a global optimization scheme, what is a solid contribution. If is clearly shown that the PID apporach is outperfomend, ad errors have lower amplitudes when disturbance rejection is concerned.
Please explain to what extent the model mismatching may affect the overall performance, as you actually use good models of your laboratory setip. What about the computational load connected to GAs? Can you use it in real-time implementation?
It would also be expected to give any information about the future directions, as any hunch about online mechanisms ensuring adaptation including RL apporoach with ADRC? Or any plans to optimize the computational burden and use edge devices, allowing space applications or UAV ones?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsPlease see the attachment.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsAccept in present form.
