Research on the SSIDM Modeling Mechanism for Equivalent Driver’s Behavior

Round 1

Reviewer 1 Report

It can be concluded that te autors used modern tools and performed modern measurements. Well-documented and precisely structured tests were performed. In connection with their results, they also presented a statistical analysis. Validation was also performed, which yielded good results.Minor deficiencies can only be observed in connection with the IDM modeling, which can be replaced with appropriate references. E.G. the following, on the theory of overtaking:

Péter, T ✉ ; Háry, A ; Szauter, F ; Szabó, K ; Vadvári, T ; Lakatos, I ✉Mathematical Description of the Universal IDM - some Comments and Application; ACTA POLYTECHNICA HUNGARICA 20: 7 pp. 99-115., 17 p. (2023) and for following the trajectories:

and for folowing the trajectories:

Peter T, Szauter F, Rozsas Z, Lakatos I Integrated application of network traffic and intelligent driver models in the test laboratory analysis of autonomus vehicles and electric vehicles; INTERNATIONAL JOURNAL OF HEAVY VEHICLE SYSTEMS  27:1-2 pp.227-245. 18. (202) with reference to Article.

 

 

 

The quality of the English language is good.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

the article is interest.

it is recommended to increase the size of the photos on the figures

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

To resolve the issue of switching boundaries between car-following and lane-changing behaviors, a failure point in the IDM, the intention to change lanes of the ego vehicle was taken into account, provided there was sufficient space and speed advantage for lane-changing. In cases where there were no benefits to changing lanes, the IDM was optimized by integrating the constraint components of the target lane vehicles and the ego vehicle's current motion state. This led to the development of the Stepless Switching Intelligent Driver Model (SSIDM) in this paper. Using the drivers' natural driving behavior data collected, an elliptic equation was employed to define the switching boundary between behaviors accurately, and the balance coefficients for the front and rear vehicles on the target lane were determined. In conclusion, the SSIDM enables seamless and precise transitions between car-following and lane-changing behaviors, surpassing the capabilities of the IDM.

Overall, this is an interesting and well-written paper. The following issue needs to be carefully addressed. 

1. The introductory literature survey lacks comprehensiveness and should include more contextual information, particularly regarding the car-following model. It is recommended to incorporate additional up-to-date references such as [1] (DOI: 10.1016/j.trb.2022.09.007), [2] (DOI: 10.1109/TITS.2021.3113788), and [3] (DOI: 10.3390/su14127045), among others. These references will contribute to a more thorough examination of the subject matter.

2. Further elaboration is necessary to explain the rationale behind selecting the Genetic Algorithm (GA) as the calibration method for the model. It is important to investigate that recent publications have introduced more advanced optimization solvers for model calibration.

3. To further enhance the comprehensiveness of the Model Validation Results section, it is advisable to incorporate additional discussions and analyses. Additionally, it would be beneficial to summarize the demographics (age, gender, annual driving mileage, etc.) of the 20 human subjects involved in the study, either in the form of a table or a graphical representation. This summary would provide a clear overview of the sample's diversity and help establish the context of the validation process.

 

4. The writing quality of this paper needs further polishing.

Further polishing of the language is required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The author worked on the development of a new algorithm for solving the problem of smooth switching between car-following model and lane changing model. The author improved the classical intelligent driver model (IDM) by using drivers' natural driving data. The scenario mining was done and Genetic Algorithm was used for identification and calibrated for the new designed model. 4 type of scenarios were tested and the performance comparisons of IDM and new designed model (SSIDM) were done. The results show that the new algorithm (model) show better performance (lane change and car following) than traditional IDM model. This paper is well written but the paper needs minor revision. The necessary corrections are listed as follows. 

1. The abstract was well-written but the author should mention the contribution of this study briefly. 

2. Please rearrange the sizes of the schemes in the flow diagram for improving the readability of the figure 5.

3. Increase the quality (resolution) of the graphs in the Figure 6.

4. The author should give more information about the physical meaning of Figure 7. What do these distributions mean? 

5. Please give more information about used Genetic Algorithm Structure and its parameters. 

6. In Figure 10, in some period of the time (for example between 150-180 sec) the IDM shows better performance than SSIDM. What are the reason of this behavior?

7. Figures 10-12 needs more explanations. The author should give more details about these results. For example, some parts of the graph in Figure 11 (195 sec-210sec) IDM and SSIDM show opposite performance, why?

8. What is the reason of the margin between the comparison results of second type, third type and fourth type scenarios in Table 5? For example, both IDM and SSIDM show better performances for type 2 and type 4 than type 3. Why? Also, the margin for IDM and SSIDM comparison for Type 3 scenario is higher than the others? Please mention these results of the Figures 10-15 and Table 5 for improving the understandability and readability of the results. 

9. The conclusion section needs to rewrite. The author should mention the details briefly by using following list. 

9.a) The goal - main purpose of the study. 

9.b) The key contributions of the study. 

9.c) The used techniques in the study.

9.d) Key findings.  

The English level is good. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 5 Report

The introduction should be supplemented. There is quite a lot of research related to the research of drivers' behavior, hence it is worth indicating the modernity of the presented model based on world publications. How to present the method of conducting research without major reservations. It is necessary to present the parameters of the script in graphic form. In this way, the presented parameters from equation (1) in the drawing would be easier to define than in the text. It is worth explaining the markings used.

Figure 2 should describe the drawings as a) and b). In Figure 2a, an author writes about how distance. In the field of lateral position? And in Figure 2b it should be rather on the axis change of speeds. Why was the beginning of the analyzed saturation not carried out at the constant value of the vehicle speed but during acceleration? In Table 1 No units. In the signature of Figure 6, no explanation a) b) c). Necessary to explain in the text (a) and (b) in Formula 12. Necessary to explain in the text (a), (b), and (c) in Formula 13. In Chapter 5.2, in the text in relation to drawings 10-12, it is worth indicating the values (errors) between the values obtained.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The revision looks solid, and the paper can be accepted without further changes.