Construction of Human Behavior Cognitive Map for Robots

Round 1

Reviewer 1 Report

This paper deals with the construction of human behavior cognitive map for robots.

The paper is interesting and easy to read. The content of the paper is suitable to the journal.

Some minor comments should be considered to improve the quality of the manuscript.

The Introduction should be enriched, since it does not provide a complete overview on the state of the art of the topic. I suggest the authors to include and comment more references about similar works that deal with human behavior and pattern recognition. The main contributions of the paper are not clear. I suggest the authors to better highlight the contributions of the paper with respect to the present literature. Furthermore, the authors should discuss the pros and cons of the proposed approach with respect to similar works already published in the literature. Figure 11 is not clear. Please describe it better. Furthermore, the figure is grainy and it should be replaced with another one with higher resolution. Some more details about the mobile robot mentioned in Section 4.3 should be included in the manuscript. Some future developments and improvements of this work should be included in conclusions.

Author Response

Dear Reviewer:

 

Thank you for the comments concerning our manuscript entitled “Construction of Human Behavior Cognitive Map for Robots” (ID: applsci-638213). Those comments are all valuable and very helpful for revising and improving our paper. We revised the manuscript in accordance with the comments, and carefully proof-read the manuscript to minimize typographical, grammatical, and bibliographical errors. Revised portions are marked in the paper.

Here below is our description on revision according to the comments.

 

Reviewer 1.

The Introduction should be enriched, since it does not provide a complete overview on the state of the art of the topic. I suggest the authors to include and comment more references about similar works that deal with human behavior and pattern recognition.

The author’s reply: Thank you for your suggestion. As suggested by the reviewer, we add some paragraphs to describe similar works on the state of the arts in lines 36 to 76.

The main contributions of the paper are not clear. I suggest the authors to better highlight the contributions of the paper with respect to the present literature. Furthermore, the authors should discuss the pros and cons of the proposed approach with respect to similar works already published in the literature.

The author’s reply: Thank you for your suggestion. We have emphasized the contribution of the behavior cognitive map and made comparison with the literature.

Figure 11 is not clear. Please describe it better. Furthermore, the figure is grainy and it should be replaced with another one with higher resolution.

The author’s reply: We add the explanation to Figure 11. And Figure 11 is replaced by a higher resolution figure. The paragraph is in lines 438 to 442.

Some more details about the mobile robot mentioned in Section 4.3 should be included in the manuscript.

The author’s reply: Thank you for your suggestion. We added a paragraph to describe the mobile robot in Section 4.3.

Some future developments and improvements of this work should be included in conclusions. 

The author’s reply: Thank you for providing these insights. We add the future developments and improvements in the conclusion.

 

Again, thank you for giving us the opportunity to revise our manuscript with your valuable comments and queries. Many grammatical or typographical errors have been revised. All the lines and pages indicated above are in the revised manuscript.

Thank you and all the reviewers for the kind advice.

 

Sincerely,

 

Wei-Zhi Lin

Author Response File: Author Response.pdf

Reviewer 2 Report

Interesting contribution. Paper is well written and methodology is appropriated.

There  are some remarks:

It could be interesting to revise Fig.1 and to complete it with labels to GRU/DHN and LMA.

Regarding the explanations about GRU vs GRU-DHN, please, explain it with more clarity (also within the experiments, figure 11).

Regarding similarities, it could be necesary to explain a little more (section 2.3).

Regarding high-level vs low-level behaviours, please explain if your approach could be extended to high-level ?.

Regarding experiments, explain better figure 13.

Regarding conclusion. This section needs to be revised, it is important to show clearer the work results, and to highlight future developments/directions, and possible applications (not only the technical details of the work, but the implications).

 

Author Response

Dear Reviewer:

 

Thank you for the comments concerning our manuscript entitled “Construction of Human Behavior Cognitive Map for Robots” (ID: applsci-638213). Those comments are all valuable and very helpful for revising and improving our paper. We revised the manuscript in accordance with the comments, and carefully proof-read the manuscript to minimize typographical, grammatical, and bibliographical errors. Revised portions are marked in the paper.

Here below is our description on revision according to the reviewers’ comments.

 

Reviewer 2.

It could be interesting to revise Fig.1 and to complete it with labels to GRU/DHN and LMA.

The author’s reply: Thanks for your suggestion. We add labels to Fig. 1.

Regarding the explanations about GRU vs GRU-DHN, please, explain it with more clarity (also within the experiments, figure 11).

The author’s reply: GRU is a structure which has the gating mechanism. It is used in recurrent neural networks to learn temporal characteristics. The difference between GRU and GRU-DHN is that GRU-DHN uses the GRU and can automatically adjust the number of neurons. However, GRU cannot automatically adjust the number of neurons.

In our method, we compared the DHN with GRU-DHN. We add sentences in lines 200-201 and 209-210 to describe the difference between DHN and GRU-DHN. We also add a paragraph to explain Figure 11 in lines 483-442.

Regarding similarities, it could be necessary to explain a little more (section 2.3).

The author’s reply: Thanks for your suggestion. We add some descriptions about the similarities in lines 270 to 276.

Regarding high-level vs low-level behaviors, please explain if your approach could be extended to high-level?

The author’s reply: The question is very interesting. The high-level behaviors are integrated/combined by a series of low-level behaviors. Therefore, we can extend our approach to high-level behaviors by recording more categories of low-level behavior cognitive maps. By combining each low-level behavior, high-level behavior can be recognized.

Regarding experiments, explain better figure 13.

The author’s reply: Thanks for your suggestion. We modify our explanation of Figure 13 from lines 480-498.

Regarding conclusion. This section needs to be revised, it is important to show clearer the work results, and to highlight future developments/directions, and possible applications (not only the technical details of the work, but the implications).

The author’s reply: We add a paragraph to show the work results and contribution. We also add the future developments and improvements in the conclusion.

 

Again, thank you for giving us the opportunity to revise our manuscript with your valuable comments and queries. Many grammatical or typographical errors have been revised. All the lines and pages indicated above are in the revised manuscript.

Thank you and all the reviewers for the kind advice.

 

Sincerely,

 

 

Wei-Zhi Lin

Author Response File: Author Response.pdf