A Review of Deep Space Image-Based Navigation Methods

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper describes the time evolution of the image navigation for deep space applications, considering the Moon, Mars and other celestial bodies. The topic is of interest to the community, but some improvements have to be made. In the following, some modifications and suggestions are to be implemented.

Introduction: Please rewrite the introduction between lines 50  and 60. As for instance, the content between the sentence at line 51 seems to contradict statements at line 57 unless these refer to celestial bodies, please rewrite clarifying. In a similar way, at line 63, the sentence should be related to what is written at line 52; even in this case, it seems that there is a contradiction.  

 

Please consider removing text from line 85 to line 89, since the preamble has already been done in the introduction.

Please add accuracy (size, resolution, landing accuracy) in Table 1 whenever possible.

Table 1 misses technical information (data rate, camera resolution, power, e.g.); these should be added.

Add loop time for NTF algorithm and related accuracy.

Please check if the acronym TM is defined before its usage (line 338).

Missing figure number at line 382.

Lines from 443 to 446 shall be removed, since the text is a repetition of the introduction.

Text from 454 to 456 should be rewritten (given the contradiction of the main hypothesis) as requested for the introduction.

Please correct Figure 2 since in the text, there are the symbols of the paragraph of ppt.

The section about AI methods for navigation shall be summarised as it is not the core of the work.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript is a comprehensive review of deep-space image-based navigation methods. The authors made an attempt to summarize and systematize existing methods used for deep space navigation mainly focusing on image recognition with the natural accent on craters analysis. Significant part of the manuscript is devoted to lunar surface recognition and different methods used in different space missions. I guess that such review might be useful for community and deserved publication. There are several point which I would like to discuss with the authors before publication.

1) After reading the manuscript I got an impression that the main focus is done on craters image analysis. On the other hand, the title “Deep space image based navigation methods” may be interpreted as  well as extended use of star trackers systems, I recommend authors either to correct the title or to make stronger accent in the abstract and  introduction

2) similar comment. the planetary missions listed in line 27 are the ones where image-based navigation is feasible. But it becomes clear only after reading the further text. Otherwise one can be surprised by the absence of Venus missions and many others.

3) Any navigation means accuracy. What I missed after reading this review is some comparison (maybe a table)  of image-based accuracy (e.g. for Lunar orbital navigation) and common ones. Typically, in many of the cited literature  these data is available (of course for certain conditions ). I understand that it may be difficult to summarize, but it is typically one of the goals to write the review. It is hard to compare methods and hard to conclude how competitive they are with earlier ones. I strongly recommend authors to add more data concerning this issue, preferably in systematic form (plots or tables).

 

Otherwise the review is quite helpful for the reader (especially for the specialists) and can be published.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript introduces a Review of Deep Space Image-Based Navigation Methods to overcome technical challenges. The content is quite rich with analysis from challenges to solutions. However, there are some issues to be further improved as well to facilitate further publication.

1. The Tracking and Control Systems Problem have less mentioned in the article. Please add the appropriate analysis;
2. Methods are mainly introduced but little analysis of solutions is done, such as Model Training and Optimization, etc. Please provide the additional analysis;
3. Section 4 “Discussion and Future Directions” must be extended with the author's own analysis;
4. Although the references in the work are famous but 35/146 references are out of date (before 2015). Please add more appropriate recent references. In addition, this article almost focuses on Deep Learning technique. However it can be seen that different Learning techniques have been considered in many previous references, such as Reinforcement Learning, Q learning, Model Free Learning, etc. https://www.sciencedirect.com/science/article/abs/pii/S0921889025002088 , https://link.springer.com/article/10.1007/s10846-024-02198-w ,

 etc. Please give some comparisons to highlight the contribution.

Comments on the Quality of English Language

Many sentences in this submission are short, such as in lines 119, 126, etc. They should be improved

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript has been generally modified as suggested. 

Best regards