OntoSLAM: An Ontology for Representing Location and Simultaneous Mapping Information for Autonomous Robots

Round 1

Reviewer 1 Report

This paper presents the concept of OntoSLAM, a model for modeling all aspects related to autonomous robots and SLAM problems for the standardization needed in robotics. I like this work.

Here are some concerns to be addressed.

In chapter 4(OntoSLAM Evaluation), you compared the similarities in lex and structure between different ontologies. It was concluded that the percentage of lexical and structural similarity was maintained at 54% and 29%. Does this conclusion prove the superiority you mentioned in the abstract? (We perform a comparative evaluation of OntoSLAM with state-of-the-art SLAM ontologies. Results show the superiority of OntoSLAM at Lexical, Structural, and Domain Knowledge levels.)

In Table 5, What are the results of the questionnaires for the different ontologies in each category? There is only one total data in table 5. In row 376, you mention " KnowRob has better performance, since it completely models the fourth category, called Workspace and the Environment Mapping categories with 87.5%", but there is no relevant data in table 5.

In line 299, you propose to use the SLAM knowledge categorization proposed in Section 2 as the gold standard for applying the evaluation method. But why is this gold standard only applied to the 4.1.3. Domain Knowledge Level?

The work is sound as claimed in the manuscript, but the presentation and discussion of the refereed data lack clarity. I would recommend the authors revise the manuscript accordingly before resubmitting.

Author Response

We appreciate your comments and suggestions. They really are helpful to improve our research. We did our best effort to take all them into consideration in the new version of our paper. We summarize the modifications done in response to your comments in blue:

In chapter 4(OntoSLAM Evaluation), you compared the similarities in lex and structure between different ontologies. It was concluded that the percentage of lexical and structural similarity was maintained at 54% and 29%. Does this conclusion prove the superiority you mentioned in the abstract? (We perform a comparative evaluation of OntoSLAM with state-of-the-art SLAM ontologies. Results show the superiority of OntoSLAM at Lexical, Structural, and Domain Knowledge levels.)

You are right that those expressions about OntoSLAM superiority are not correct. To clarify that, we have modified the abstract and the introduction to clearly point out that OntoSLAM is superior in Domain Knowledge level, but similar at Lexical and Structural levels. We also have added sentences in Section 4.1.3 and Section 4.1.4 to reflect that. In Section 4.3, there is the explanation of the comparative evaluation of OntoSLAM and other SLAM ontologies that was more certain.

In Table 5, What are the results of the questionnaires for the different ontologies in each category? There is only one total data in table 5. In row 376, you mention " KnowRob has better performance, since it completely models the fourth category, called Workspace and the Environment Mapping categories with 87.5%", but there is no relevant data in table 5.

To improve this aspect, we explain better Table 5 and show how the ontologies are compared based on the Domain of Knowledge with the metric of percentage of questions answered.

In line 299, you propose to use the SLAM knowledge categorization proposed in Section 2 as the gold standard for applying the evaluation method. But why is this gold standard only applied to the 4.1.3. Domain Knowledge Level?

The comparison against a golden-standard (GS) is only possible if it is available as an ontology. In our case, the GS is a categorization of SLAM knowledge. We better explain this aspects in sections 4.1.1, 4.1.2, and 4.1.3.

The work is sound as claimed in the manuscript, but the presentation and discussion of the refereed data lack clarity. I would recommend the authors revise the manuscript accordingly before resubmitting.

Thanks again for the suggestions. We did our best effort to reach a better quality of the manuscript and improve the explanations. All the modifications performed are marked in red in the new version of the manuscript. 

Author Response File: Author Response.pdf

Reviewer 2 Report

1. Navigation of autonomous vehicles is one of the most important research topics in recent years. One of the leading technologies in this field is SLAM technology. The reviewed article is devoted to this topic.
2. General remarks
   1. Please clearly state the scientific objective of the paper and the elements of novelty in relation to known and used SLAM methodologies.
   2. Please use the language of a scientific research report without personal references: line 164, “we propose OntoSLAM”, line 188 “we follow a three-step methodological”, line 205 “we use the, line 432 “we implemented the following” and many others in whole article.
   3. The paper lacks information about the data sources used for processing with SLAM.
   4. However the article is very well written should be carefully edited. Very few remarks included below.
3. Specific remarks
   1. I suggest describing more extensively the experiment conducted.
   2. The final conclusions are too general and only generally summarize the research presented in the article. I suggest expanding the conclusions with more detailed findings.
   3. Only initials are used in the chapter outlining the authors' contributions. This chapter currently occupies 11 lines.

Author Response

We appreciate your comments and suggestions. They really are helpful to improve our research. We did our best effort to take all them into consideration in the new version of our paper. We summarize the modifications done in response to your comments in blue:

• General remarks
  • Please clearly state the scientific objective of the paper and the elements of novelty in relation to known and used SLAM methodologies. 

We have modified the abstract, introduction, and related work sections to explicitly state the contribution of our work.

• • Please use the language of a scientific research report without personal references: line 164, “we propose OntoSLAM”, line 188 “we follow a three-step methodological”, line 205 “we use the, line 432 “we implemented the following” and many others in whole article.

Although, we do not agree that our writing style is not appropriate for research documents, we have changed to an impersonal redaction, as the reviewer asked. We consider that the impersonal redaction is just another rigth writing style, but not the only one.

1.  
   • The paper lacks information about the data sources used for processing with SLAM.

We added a description of the source of the information used for the SLAM, both for the Gmapping algorithm and octomap mapping, in Section 4.2.4.

• • However the article is very well written should be carefully edited. Very few remarks included below.

We carefully read several times the paper to improve and correct the writing style.

1. Specific remarks
   • I suggest describing more extensively the experiment conducted.

We specified more details about the platform used for the experiments and about the classical SLAM algorithms executed in Section 4.2.4.

• • The final conclusions are too general and only generally summarize the research presented in the article. I suggest expanding the conclusions with more detailed findings.

We changed the conclusions focusing on the most important findings and future research

• • Only initials are used in the chapter outlining the authors' contributions. This chapter currently occupies 11 lines.

Done.

Thanks again for the suggestions. We did our best effort to reach a better quality of the manuscript and improve the explanations. All the modifications performed are marked in red in the new version of the manuscript. 

Author Response File: Author Response.pdf

Reviewer 3 Report

This manuscript addresses an ontology to model the aspects related to autonomous robots and the SLAM problem, modelling the dynamics of the SLAM process by including uncertainty of robot and landmarks. It is an interesting topic for the community

It is well written, and results are convincing.

References 6 and 7 should be more detailed (what is the background and improvements).

Figure 7 resolution and quality is different from the others. Please replace it accordingly.

Minor typo: Line 48: show -> shows

Author Response

We appreciate your comments and suggestions. They really are helpful to improve our research. We did our best effort to take all them into consideration in the new version of our paper. We summarize the modifications done in response to your comments in blue:

References 6 and 7 should be more detailed (what is the background and improvements).

A description of both references and their relationship to the manuscript were added in Section 1.

Figure 7 resolution and quality is different from the others. Please replace it accordingly.

The quality of the images was already the same, however, to keep consistency, the proportion of Figure 7 was modified.

Minor typo: Line 48: show -> shows

Done.

Thanks again for the suggestions. We did our best effort to reach a better quality of the manuscript and improve the explanations. All the modifications performed are marked in red in the new version of the manuscript. 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have revised the paper in accordance with the reviewers instructions and the paper can now be published.