Recognition of Lane Changing Maneuvers for Vehicle Driving Safety

Round 1

Reviewer 1 Report

The study is based on the analysis of driving trajectories focusing on two basic driving manoeuvres on highways lane keeping and lane changing.

The paper has some critical issues.

1. The literature review could be better. Authors should add studies related to driving trajectories (doi:10.3390/su10114248; DOI: 10.7708/ijtte.2016.6(4).06)

2. The description of the data. The input variables used are not clear. Furthermore, the authors should include summary information on the driving profiles, such as the mean and standard deviation of the various measurements, possibly adding figures showing the most representative driving profiles of the different driving scenarios.

3. In section 3.2, the description of the SVM algorithm and HMM algorithm should be removed. Only the description of the K nearest Neighbor algorithm needs to be reported.

4. In the HighD dataset paragraph, authors must specify whether all recorded data has been used or vehicles have been removed. Furthermore, being a naturalistic study, the authors also need to specify the boundary conditions, such as climatic and traffic conditions, if any ve

Author Response

Response to Reviewer 1 Comments

Point1: The literature review could be better. Authors should add studies related to driving trajectories (doi:10.3390/su10114248; DOI: 10.7708/ijtte.2016.6(4).06)

Response 1: The studies related to driving trajectories has been added in the article thanks to the recommandation.

Point 2: The description of the data. The input variables used are not clear. Furthermore, the authors should include summary information on the driving profiles, such as the mean and standard deviation of the various measurements, possibly adding figures showing the most representative driving profiles of the different driving scenarios.

Response 2: The authors make the input variables more clear. Driving profiles are also been mentioned in the article.

Point3: In section 3.2, the description of the SVM algorithm and HMM algorithm should be removed. Only the description of the K nearest Neighbor algorithm needs to be reported.

Response 3: The description of the SVM algorithm and HMM algorithm are removed from article thanks to the recommandation.

Point 4: In the HighD dataset paragraph, authors must specify whether all recorded data has been used or vehicles have been removed. Furthermore, being a naturalistic study, the authors also need to specify the boundary conditions, such as climatic and traffic conditions, if any ve

Response 4: The authors specify the certain recorded data used in study. However, there is no boundary conditions such as climatic or traffic conditions recorded in the dataset. In that case, the authors are able to specify it.

All the changes can be seem in the attachment files. They are all in red.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors study Lane keeping and lane changing using of HighD trajectory dataset extracted by drone videography.

The literature analysis is scarce; for example, the authors talk about lane keeping (LK), they should insert examples of reference studies specifying the definition of lane keeping ( Galante et al. [DOI: 10.1177/03611981211034718] studied lane keeping analyzing the Mean and Standard Deviation of Lateral Position [DOI: 10.1061/(ASCE)0733-947X(2006)132:8(649)] defined Deviation of Lateral Placement (DLP) as an index representing a driver's steering behaviour along a given section of a highway) and examples of lane changing ( Brijs et al. [Doi: 10.1016/j.aap.2022.106763 ] studied lane variation by analyzing the lateral position and Vehicle heading angle).

Data extraction is unclear. The authors should specify how extracted the data used for analysis.

Authors must enter a summary table of the input data used for the KNN analysis.

To make the reading of the paper clearer, paragraph 4 must be moved before paragraph 3 and changing the name to, for example, a driving scenario.

Authors should describe the Skewness of acceleration values for groups, Lane Keeping(LK) and Lane Changing(LC).

Author Response

Response to Reviewer 2 Comments

Point 1: The literature analysis is scarce; for example, the authors talk about lane keeping (LK), they should insert examples of reference studies specifying the definition of lane keeping ( Galante et al. [DOI: 10.1177/03611981211034718] studied lane keeping analyzing the Mean and Standard Deviation of Lateral Position [DOI: 10.1061/(ASCE)0733-947X(2006)132:8(649)] defined Deviation of Lateral Placement (DLP) as an index representing a driver's steering behaviour along a given section of a highway) and examples of lane changing ( Brijs et al. [Doi: 10.1016/j.aap.2022.106763 ] studied lane variation by analyzing the lateral position and Vehicle heading angle).

Response 1: The authors add the studies related to lane keeping and lane changing in the literature review thanks to the recommandation.

Point 2:Data extraction is unclear. The authors should specify how extracted the data used for analysis.

Response 2: The data extraction in the article is more clear.

Point 3: Authors must enter a summary table of the input data used for the KNN analysis.

Response 3: Table is added in the article.

Point 4: To make the reading of the paper clearer, paragraph 4 must be moved before paragraph 3 and changing the name to, for example, a driving scenario.

Response 4: The paragraph 4 is moved before paragraph 3 and changing the name to Driving scenario.

Point 5: Authors should describe the Skewness of acceleration values for groups, Lane Keeping(LK) and Lane Changing(LC).

Response 5: Authors describe the Skewness of acceleration values for groups, Lane Keeping(LK) and Lane Changing(LC) in the article thanks to the recommandation.

All the changes can be seen in the attachment file. They are all in blue.

Author Response File: Author Response.pdf

Reviewer 3 Report

Review

 

Recognition of Lane-Changing Maneuvers for Vehicle’s Driving Safety

 

Yuming Wu and Lei Zhang

 

Subject

The research topic is interesting and appropriate for the journal requirements.

 

Title

Clear and corresponds to the content of the article.

 

Abstract

Appropriate, including the purpose of the research. The abstract is clear, arouses interest, and consists of the goals.

 

1. Introduction

The introduction is of sufficient scope.

 

2. Literature review

The presentation of previous studies is appropriate. The number of the earlier studies presented is good as there are previously known scientific publications on the subject.

 

3. Methodology

The presentation of the method is accurate. The description of the method is appropriate and understandable.

 

4. HighD dataset

The HighD trajectory dataset appropriate.

 

5. Result

The presentation of the results is understandable and well presented. The figures are good for understanding the study. In figure 7, I recommend using transparent circles.

 

6. Conclusion

The conclusions are appropriate based on the results presented.

 

References

The references in the literature presented are well suited to the topic. In addition, the quality, quantity, and value of the cited papers are excellent.

 

Overall merit

This paper is well structured and sufficiently detailed. The descriptions are understandable; the results are well presented and evaluated. 

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 3 Comments

Thanks for your kindness.

And we change the figure you mentioned by using more transparent circles.

You can see it in the article, the new figure is figure 8.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

the authors responded to all reviews

Reviewer 2 Report

the authors responded to all reviews