Collision Avoidance Behavior Mining Model Considering Encounter Scenarios

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

- It would be good to see in further researches how much developed recognition algorithm is reliable depending on other circumstances or under particular sea areas such as narrow straits/ port basins etc.....

- Would be good to see how the algorithm behave if some other data been used out of AIS equipment.

-Path planning algorithm for autonomous ships (how much influence will be from different size of autonomous ships)

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, I suggest a few changes to make the material clearer and easier to understand in certain areas. 

 INTRODUCTION

1.      I suggest completing the sentence: “However, no specific quantitative criteria”

METHODOLOGY

2.      Please provide a proper justification for setting the criteria for potential collision risk as DCPA less than 1.5 nautical miles and TCPA greater than 0 but less than 30 minutes, other than “It is generally believed that when the DCPA is less than 2 nautical miles and the TCPA is greater than 0, there is a potential collision risk between ships”

3.      Equation (10) appears twice. The equations need to be renumbered.

RESULTS

4.      I suggest proposing another title for the paragraph “Comparative Experiment”

5.       Please clarify the term "timestamp".

6.      Please provide appropriate units of measurement for the axes of all graphical representations.

CONCLUSION

7.      Please, clearly highlight the proposed model’s effectiveness in accurately identifying collision avoidance behaviors and encounter scenarios, thereby providing valuable data for autonomous navigation and intelligent waterway traffic management.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Manuscript title: Collision Avoidance Behavior Mining Model Considering En-2 counter

Scenarios (applsci-3422928 )

The manuscript presents an interesting algorithm for the computation and possible automation of a sequence of maneuvers to be conducted in order for a ship to avoid collision with other ships. This reviewer believes that the manuscript would merit publication after a few simple changes are made. These proposed changes, will improve only a little bit on the manuscript presentation and clarity. Authors have done a nice work on the preparation of the manuscript.

1.- Authors please explain the role of index “i” in equation (5) page 10. It is not clear if the index “i” fulfils only an iterative purpose on the algorithm or if it is also associated to a time sample index. Given that the ship is in motion, this reviewer assume the index “i” points to data that was acquired for an specific position of the ship, at an specific time. Please clarify.

2.- In page 11, line 357, an apparent misplaced sentence reads “(1) Ship Collision Avoidance Turning Point Recognition”. Is the sentence actually out of place? If so, please delete the sentence.

3.- Figures 5 in page 11 and Figure 6 in page 13 both illustrates the sliding window algorithm. Could authors please indicate if the motion of the ship as illustrated corresponds to the motion as measured or observed from the center of the ship? Please add a brief sentence indicating if the center of the ship is the one depicted in Figure 5 as a small empty circle.

4.- Equation (13) in page 15 is written using slanted or italic letters for the function sin. This is not the usual way of writing equations. Authors please write the function “sin” in Equation (13) using regular letters.

Author Response

Please see the attachment

Author Response File: Author Response.pdf