Review Reports

Round 1

Reviewer 1 Report

The paper presents an overview of the geometric design standard for freeway and how it is possible to adapt it in autonomous driving environment.

I think that this study is superficial and not founded on scientific field. In fact, only on horizontal geometric design there are an extensive literature (for example more studies by Cafiso et al. talked about the importance of curvature radius in autonomous  driving). 

For this reason, I propose to the author to rewrite the paper basing on scientific basis.

 

Author Response

Response to Reviewer 1 Comments:

Point 1: I propose to the author to rewrite the paper basing on scientific basis.

Response 1: In this revision, we improved the structure of the article, enriched the content of the introduction section, and added a literature review section. At the same time, we analyze and compare the traditional driving environment and the autonomous driving environment to improve the scientificity of the article. And added references to some assumptions and ideas in the article. We discussed in depth the principle of many design indicators changing in the autonomous driving environment，and highlighted their impact on autonomous vehicle.

Author Response File: Author Response.docx

Reviewer 2 Report

The introduction of automated vehicles in the market will likely have a great impact on the road and transport design, management, operation and safety. This paper focuses on the geometric design standard for freeway under fully autonomous driving environment. The topic is interesting because it can provide basic insight on how autonomous driving could change road geometric design. However, there are some important issues in this manuscript. First, this manuscript is not well-structured manner. Second, the research deep is not enough for an article.

Some additional comments were listed below:

1.     Introduction section shows 4 main components: general background, specific background, knowledge gap and ‘here we show…’. However, the introduction provided is not sufficient with lack of relevant references. A statement of study objective, a brief summary of the study and the implication/novelty of the results can be explained at the end of the section.

2.     The aim/objective of the study is not clear.

3.     Figure 1 shows the transformation of the system. However, both before and after transformation represent autonomous driving. Usually, traditional transportation system consists of 3 components: human/driver, vehicle and road environment. Road and network are believed under the same component.

4.     Section 2 and section 3 provide relevant literature related to the area of research. However, (i) the writing flow must be improved. These sections also contain objective and methodology, (ii) lack of literature review on the impact of autonomous vehicles on geometric design. Highlight key characteristics and the impact of those characteristics on guidelines for road geometric design due to transition from the non-autonomous driving to autonomous driving.

5.  Figure 2 and Figure 3 are not well explained.

 

5.     Discussion section tells the key findings. However, the review in previous sections is not comprehensive.

Author Response

Response to Reviewer 2 Comments:

Point 1: Introduction section shows 4 main components: general background, specific background, knowledge gap and ‘here we show…’. However, the introduction provided is not sufficient with lack of relevant references. A statement of study objective, a brief summary of the study and the implication/novelty of the results can be explained at the end of the section.

Response 1: Introduction. Line 41-88. Added the introduction to the traditional artificial driving environment and the autonomous driving environment. Added relevant references. Line 106-113. Added a brief explanation of the significance and role of the findings.

Point 2: The aim/objective of the study is not clear.

Response 2: Introduction. Line 103-113. Added description of research objectives

Point 3: Figure 1 shows the transformation of the system. However, both before and after transformation represent autonomous driving. Usually, traditional transportation system consists of 3 components: human/driver, vehicle and road environment. Road and network are believed under the same component.

Response 3: Line 27-31. Modified the figure and its description to make it more in line with people's cognition and avoid ambiguity.

Point 4: Section 2 and section 3 provide relevant literature related to the area of research. However, (i) the writing flow must be improved. These sections also contain objective and methodology, (ii) lack of literature review on the impact of autonomous vehicles on geometric design. Highlight key characteristics and the impact of those characteristics on guidelines for road geometric design due to transition from the non-autonomous driving to autonomous driving.

Response 4: (i) Line 199-204. Line 426-434. Added objectives and methods of these two sections to make sections‘ structure more complete. (ii) Literature Review. Line114-197. Added literature review on the development of autonomous driving technology and the impact of autonomous vehicles on geometric design.

Point 5: Figure 2 and Figure 3 are not well explained.

Response 5: Line 237-253. The braking process of the vehicle in traditional artificial driving environment is explained in more detail. And explained the meaning of the symbol in figures.

Point 6: Discussion section tells the key findings. However, the review in previous sections is not comprehensive.

Response 6: Line 914-919. Line 930-955. The discussion section is enriched and the findings of the previous sections are reviewed. Supplements the missing section and briefly discusses the advantages of autonomous vehicles.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper is really interesting and deals with a very current topic.

The authors explore and study the formulation principles and parameter values of some design in-dicators in traditional driving environment (in particular they refer to Chinese Design Specification for Highway Alignment). At the same time, they analyze the compatibility and adaptability of the corresponding indicators in the autonomous driving environment. Then, they propose how to improve the geometric design standards of highways, make them more suited to the traffic characteristics of autonomous vehicles, and achieve coordination between the needs of autonomous vehicles and the design of highways.

I would like to receive some clarifications from the authors:

1. line 309: HOW DO THE AUTHORS CALCULATE  THESE LATERAL FORCE COEFFICIENTS?

2. lines 330-339: COULD THE AUTHORS EXPLAIN WHERE THIS CLASSIFICATION COMES FROM?

3. lines 420-422: COULD THE AUTHORS EXPLICIT WHY THEY ASSUMED THE CONDITION IN WHICH VERTICAL CURVE LENGTH IS GREATER THAN STOPPING SIGHT DISTANCE TO CALCULATE THE MINIMUM RADIUS OF THE CONVEX VERTICAL CURVE?

Author Response

Response to Reviewer 3 Comments:

Point 1: How do the authors calculate these lateral force coefficients?

Response 1: Line 483-500. Added the description of the value of the lateral force coefficient.  Fully autonomous highways have higher requirements for passenger comfort. Referring to the research by Chinese scholars，these lateral force coefficients are reasonably selected in consideration of passenger comfort requirements.

Point 2: Could the authors explain where this classification comes from?

Response 2: Line 520-526. Added the description of the classification comes from. This classification is derived from different requirements for road width at different design speeds.

Point 3: Could the authors explicit why assumed the condition in which vertical curve length is greater than stopping sight distance to calculate the minimum radius of the convex vertical curve?

Response 3: Line 647-653. Added explanation of why formula (6) is used for calculation.

Author Response File: Author Response.docx

Reviewer 4 Report

The main idea and objectives of the research:

·        The main objectives of this research is to propose a new geometric design standards for freeway in fully autonomous driving environment.

·        The autor started the research by proposing a new design speed for a passenger car at the Freeway equals to 180 km/h, considering that the characteristics of the future car will not differ from the current one. Then reconstruct the Stopping Sight Distance (SSD) equation based on the suggested design speed while ignoring the perceptual reaction time (PRT) for human influence.

·        Based on the new SSD, other design parameters have been re-built.

 

My comments:

·        What is noted is that most if not all of the new recommended design standards are controlled by the SSD equation and mainly by the new recommended design speed (180km/h), whereas the effect of environment and traffic conditions are neglected. For instance the coefficient of friction by different whether condition has been neglected in the design of horizontal curve.

·        It is recommended to have more references especially that some hypotheses and ideas in the text lack references (example Line 67) .

·        It is recommended to compare the resulting design values in the tables with the value of the current design standards.

·        Of course, excluding the human turn will reduce the SSD, but in return, will not the SSD doubled more again with the proposed high speed?, as shown in Table 5 and also the other design elements.

·        Considering that the design speed is very high 180km/h, it is important to talk about and compare safety aspects especially in case of system failure.

·        Tables (11 and 12): represents the radius of the proposed horizontal curve. Won't the long radii (2.5 and 1.7 km) affect passenger comfort?

·        Line 370: You suggest canceling the maximum horizontal curve radius.  The current standard maximum radius is suggested to maintain safety, passenger comfort, and economical and land use aspect which are not all covered in this proposal.

·        Table (14): The suggested values for the length of the vertical curve (28500 m) for the proposed high design speed are still not reasonable, especially for passenger comfort and terrain conditions. Can you comment more on that.

·        However, if the design speed will reach 180 km/h then more robust design standards will be required

Author Response

Response to Reviewer 4 Comments:

Point 1: What is noted is that most if not all of the new recommended design standards are controlled by the SSD equation and mainly by the new recommended design speed (180km/h), whereas the effect of environment and traffic conditions are neglected. For instance the coefficient of friction by different whether condition has been neglected in the design of horizontal curve.

Response 1: Line 490-492. The lateral force coefficient in the horizontal curve calculation model takes into account the influence of the friction factor between the tire and the road surface. The most unfavorable conditions are considered when the lateral force coefficient is selected to meet the needs of different situations.

Point 2: It is recommended to have more references especially that some hypotheses and ideas in the text lack references (example Line 67).

Response 2: Literature Review. Line 222. Added literature review section. And added references to some assumptions and ideas in the article.

Point 3: It is recommended to compare the resulting design values in the tables with the value of the current design standards.

Response 3: Line 545-552. Line 561-565. Line 573-578. Line 735-736. Line 794-800. Added comparison of calculated values ​​with current standard values.

Point 4: Of course, excluding the human turn will reduce the SSD, but in return, will not the SSD doubled more again with the proposed high speed?, as shown in Table 5 and also the other design elements.

Response 4: Higher design speed places higher demands on SSD. This demand is reflected in the calculation formula as a higher initial braking speed. Therefore, the effect of high design speed on SSD length has been included in the SSD calculation model.

Point 5: Considering that the design speed is very high 180km/h, it is important to talk about and compare safety aspects especially in case of system failure.

If the design speed will reach 180 km/h then more robust design standards will be required.

Response 5: First of all, I agree with your point of view. However, the autonomous driving technology is not yet complete, and there are few studies on failure mechanism of the autonomous vehicle system. High design speeds tend to lead to more serious traffic accidents, so high design speeds require stricter design standards to ensure driving safety. This paper mainly considers changes in design indicators brought by changes in driver's psychophysiological characteristics. Less consideration is given to the impact of vehicle systems on safety at high design speeds. This is the limitation of this paper. I will focus on this aspect in future research.

Point 6: Tables (11 and 12): represents the radius of the proposed horizontal curve. Won't the long radii (2.5 and 1.7 km) affect passenger comfort?

Response 6: Large-radius horizontal curves tend to have a bad effect on the driver's psychology. But fully autonomous vehicles do not have a driver, so this effect can be negligible. When the vehicle is driving on a horizontal curve with a large radius, its stress state is almost the same as that of a straight section. Therefore, the comfort of the passengers can be guaranteed on the horizontal curve with a large radius.

Point 7: You suggest canceling the maximum horizontal curve radius.  The current standard maximum radius is suggested to maintain safety, passenger comfort, and economical and land use aspect which are not all covered in this proposal.

Response 7: Line 579-589. Line 594-598. Added considerations for safety, comfort and economy.

Point 8: Table (14): The suggested values for the length of the vertical curve (28500 m) for the proposed high design speed are still not reasonable, especially for passenger comfort and terrain conditions. Can you comment more on that.

Response 8: Line 743-750. Added discussion of limit  values and general values. This article prefers to recommend the use of general values ​​for passenger comfort.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thanks to the authours to address my suggestions

Author Response

请参阅附件。

Author Response File: Author Response.docx

Reviewer 2 Report

The comments are listed below:

1. line 28: "...will transform into the "vehicle-road-cloud-network"..." The elements stated in the sentence are not consistent with the elements shown in Figure 1.

2. line 126: AV, PRT - Must be clearly stated before using the short form of the word.

3. Line 128: "....requirements of level 3 and level 5 autonomous vehicles." Suggested to use L3  for level 3 and L5 for level 5 for consistency. Please check other sentences.

4. line 254: "The driving distances in t T1, T2 and T3 are S1, S2 and S3." Suggested to revise this sentence especially when the authors use word 't' in the sentence.

5. line 325: 3.4m/s2 - Please use correct unit 3.4m/s². This comment applied to all related units.

6 line 567-572: "The minimum radius of curve without superelevation at the design speed of 80km/h, 567 100km/h and 120km/h when cross slope is less than 2% in the Chinese standard is 5500, 568 4000 and 2500, "

"........... greater than 2%, the minimum radius of curve without superelevation with the design speed of 80km/h, 100km/h and 120km/h is 7500, 5250 and 3350, ......".

In these 2 sentences, the design speed is stated from 80 km/hr up to 120 km/h. However, the minimum radius is stated from 120 km/hr to 80 km/h.  Please state the minimum radius following the design speed from 80 km/hr to 120 km/hr.

7. line 643: Please insert formula number 

8. line 730-731: "...is reduced by 1,000 m, 3,500 m and 5,000 m at 80 km/h, 100 km/h and 120 km/h design 730 speeds, as shown in Figure 4.5". There is no Figure 4.5 in the manuscript and please check 5000 m. If refer to Figure 10, the difference is 4900 m.

Author Response

Please see the attachment.

Author Response File: Author Response.docx