A Comparative Study of Indoor Accuracies Between SLAM and Static Scanners
Round 1
Reviewer 1 Report
Comments and Suggestions for Authorsapplsci-3739865-peer-review-v1
The manuscript “Testing the Indoor Accuracies of SLAM and Static Scanners” addresses an interesting topic, which adhere to Applied Sciences journal policies. The manuscript tackles a topic of interest, related to presenting a detailed and methodological comparison between static laser scanners and handheld/mobile SLAM scanners in an indoor corridor environment. Five devices, two static (Leica RTC 360, Trimble X7) and three SLAM systems (GeoSLAM ZEB Horizon RT, Emesent Hovermap ST-X, FARO Orbis), were tested for absolute and relative accuracy, noise levels, and local profile deviations using a high-precision static scan and total station network as the reference.
This is a good and well-structured manuscript with adequate technical depth, and a comprehensive testing protocol. The authors’ attention to detail and commitment to reproducibility are commendable. However, despite these strengths, the manuscript must address several weaknesses before publication.
A few medium/major improvement suggestions:
- My biggest recommendation is: You need to include some basic statistical measures like standard deviation or confidence intervals for the RMSE values shown in Tables 1-5. Right now, you're only showing averages, which doesn't tell us anything about how consistent the results were across multiple passes, especially for the SLAM devices. Just adding SD columns to those tables and maybe a simple bar chart or boxplot in the Results section would really help show the variability and strengthen the conclusions.
- The local profile evaluation relies entirely on the reference cloud and ICP, but there’s no independent ground truth to back it up. You're assuming the reference cloud is perfectly accurate everywhere, including the central part of the corridor, but that might not always be the case. It would really help to include a few control points measured with the total station or P40 targets in that area to confirm local accuracy. If that’s not possible, at least acknowledge this limitation clearly in the Discussion section.
- In Section 2.4, the Trimble X7 setup used 22 stations while the Leica RTC 360 used 19, and their point densities were also slightly different (11 mm vs. 12 mm at 10 m). But there’s no explanation for why these differences were necessary or how they might affect the results. It would help to add a short paragraph explaining the rationale behind these settings, were they based on manufacturer recommendations, practical constraints, or scanner capabilities? Also, briefly comment on whether these differences could have influenced the accuracy comparison.
- In the Conclusions section, you say SLAM scanners are "becoming viable," but it would really help to go one step further and spell out where they’re actually a good fit. Right now, there's no mention of concrete use cases.
I recommend major revisions before the manuscript is considered for publication. I also kindly ask the authors to colour the changes inside the text in the revision.
Author Response
The response to the review is in the attached file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for preparing this article. Overall, the paper is well written; I have just a few recommendations and comments:
Line 176: „It has it 3 calibrated cameras“– I believe there is a typo here; please verify.
Lines 235 and 236: The authors stated: „...58 control points on the walls and on the floor were georeferenced using the total station (from five positions) with a mean coordinate error of 0.4 mm.“ – Could you please clarify how this coordinate error was calculated?
Lines 250-251: Was the final registration performed during export in the Trimble Perspective software based solely on automatic registration, or did the authors manually influence the transformation in some way? And what was the accuracy of this transformation?
Lines 289-299: Why was decimation not applied to the point clouds obtained from MLS?
The article is generally well written and provides a sufficient amount of information, which is supported by thorough testing. At the same time, it highlights the technological advancements in the field of SLAM MLS scanners.
Regards
Author Response
The response to the review is in the attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper focuses on comparing the accuracies of different generations of SLAM and static scanners in indoor measurement settings. Through systematic testing and analysis of various scanners, it offers valuable data and insights for indoor measurement technology. The study is innovative and practical, potentially advancing indoor measurement techniques. However, it also has some shortcomings.
1. The title's "testing" is too vague. A more specific title like "A Comparative Study of Indoor Accuracies between SLAM and Static Scanners" would better capture the study's essence.
2. The abstract states the research purpose and main results but doesn't highlight the innovation of comparing new and old - generation SLAM scanners. This makes the study's potential contribution to indoor measurement technology less clear. The abstract should emphasize this innovative aspect more.
3. The introduction briefly introduces SLAM technology but lacks in - depth analysis of its current development trends. It doesn't adequately cover the technology's evolution, current status, and accuracy issues in indoor surveying. This weakens the study's claimed necessity and urgency. The introduction should include a more thorough discussion of SLAM technology's development trends.
4. The transition from the introduction to the materials and methods section is a bit abrupt. The logical link between the research background and subsequent methods isn't well - established. Adding transitional content would make the connection between the sections more natural and smooth.
5. On page 3, line 116, non - standard English symbols/words like "až" appear. This could confuse readers. Replace such non - standard expressions with English words like "to" or "until" to ensure language consistency and accuracy.
6. The clarity of Figure 6 needs enhancing to ensure readers can understand the chart's content. High - definition processing of charts before formal publication would improve the document's overall quality.
7. The symbol "RMSEICP - ALL" appears on page 9, line 344, but its definition isn't provided in the document. To avoid confusion, clearly explain this symbol's meaning in the document to help readers better understand the study.
8. The results section mainly uses tables and text to present data, lacking visualizations like line graphs and bar charts. This reduces data intuitiveness and readability. Adding data visualizations would improve the results' presentation.
9. The discussion section focuses on interpreting experimental results and comparing them with existing research but doesn't adequately address the experimental design's limitations. A deeper analysis of experimental results is also needed to fully explore their potential significance and application value.
10. The conclusion has repetitive descriptions and is too lengthy. For example, it repeatedly compares the performance of static scanners (e.g., Leica RTC 360) and SLAM scanners (e.g., Emesent Hovermap ST - X, FARO Orbis, GeoSLAM ZEB Horizon RT). The conclusion should be rewritten to eliminate redundancy, making it more concise while emphasizing the study's key findings and contributions.
Author Response
The response to the review is in the attached file.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have improved the manuscript and provided responses to all reviewer suggestions. As a result, the manuscript is now methodologically stronger and more clearly presented. In my opinion, it can be considered for publication.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have answered all my questions.
