Traffic Manager Development for a Roundabout Crossed by Autonomous and Connected Vehicles Using V2I Architecture
, Stephane Camp 2,3, Charles Verhas 2, Nicolas Pollet 1,2, Benjamin Ortega 2 and Baptiste Robic 2Round 1
Reviewer 1 Report (Previous Reviewer 4)
Additional explanations by the Authors and extensions to the text of the manuscript are satisfactory.
Author Response
There were no particular comments, the paper has been generaly reviewed and enhanced.
Reviewer 2 Report (New Reviewer)
I can't understand why there are already red sentences in the paper if this is the first round of revision. moreover, on each page, the lines are not numbered so it is also difficult for me to try to make the comments for the authors easy to understand...Anyway, the authors' aim of the paper was to manage the traffic at roundabouts crossed by CAVs. CAVs represent the challenge of the moment so the study is welcome. Below I provided some comments to improve the paper and its readability:
1) abstract. please do not provide the references in the abstract.
2) page 2, in the paragraph in red, AI was used without defining it. Please, use acronyms only after having introduced them first.
3) Fig 1.
- The comments are difficult to read.
- In the roundabout vignette, "coodonate" was written instead of coordinate.
4) Materials. the authors stated "It has been assumed in this study that all the users of the network can communicate their positions to the central unit". Is this assumption realistic or not? or is this assumption something futuristic?
5) section 3.4.
- are there any references supporting your choice of doing the simulation with vehicles arriving randomly with a period of 2 seconds?
- the authors stated that if there is a collision of vehicles during the simulation, the simulation stops and a flag is lifted. This flag hasn't been observed during the simulations done with the developed traffic manager. what is your opinion about this?
- what is EGMO5? the authors just said that it is a benchmark simulator but the sentence, as it is, lacks interpretability and comprehension by the reader.
6) section 4.1
- did the authors consider any geometric standards in their study? And if this is the case, is the result compatible with the standards? I mean, many standards do not allow the use of three entry lanes (see this roundabout overview for further info dx.doi.org/10.1139/cjce-2013-0123). moreover, it is not safe to increase the number of entry lanes up to three whereas the circulating lanes remain for instance 2. please explain your choice.
7) table 5.
- Please, explain how the combinations from 1 to 8 can be identified and understood by the reader.
- also explain what do -1 or 1 value stand for.
8) Recently, there is an increasing interest in proactive road safety assessments based on the examination of the characteristics of the road to identify the presence of risk factors (i.e., https://doi.org/10.1016/j.aap.2022.106858 was carried out for roundabouts). How may your study be of support for such safety approaches? Can be the proactive approach potentially combined with your results? Please add a comment on the topic.
Minor editing of English language required
Author Response
1) abstract. please do not provide the references in the abstract.
⇒ Done
2) page 2, in the paragraph in red, AI was used without defining it. Please, use acronyms only after having introduced them first.
⇒ Done
3) Fig 1.
- The comments are difficult to read.
- In the roundabout vignette, "coodonate" was written instead of coordinate.
⇒ Done
4) Materials. the authors stated "It has been assumed in this study that all the users of the network can communicate their positions to the central unit". Is this assumption realistic or not? or is this assumption something futuristic?
⇒ Technically realistic, in practice limited by confidentiality issues
5) section 3.4.
- are there any references supporting your choice of doing the simulation with vehicles arriving randomly with a period of 2 seconds?
⇒ It is an upper-bound to the traffic flow observed
- the authors stated that if there is a collision of vehicles during the simulation, the simulation stops and a flag is lifted. This flag hasn't been observed during the simulations done with the developed traffic manager. what is your opinion about this?
⇒ We checked that when introducing an error in our code, the flag was lift thus we are sure that the flag hasn’t been lifted because no collision was observed. In addition, we are aware that this flag represent the collision due to our traffic manager. The environment tested was the simulation and we cannot guarantee that there won’t be any collision if we are outside of the assumptions made for this study : vehicle can’t communicate its position, etc..
- what is EGMO5? the authors just said that it is a benchmark simulator but the sentence, as it is, lacks interpretability and comprehension by the reader.
⇒ EGMO are the initials of one of the authors. We believe that a comment has mistakenly been included in the text. Concerning the benchmark simulator, it is further discussed in section 4.4
6) section 4.1
- did the authors consider any geometric standards in their study? And if this is the case, is the result compatible with the standards? I mean, many standards do not allow the use of three entry lanes (see this roundabout overview for further info dx.doi.org/10.1139/cjce-2013-0123). moreover, it is not safe to increase the number of entry lanes up to three whereas the circulating lanes remain for instance 2. please explain your choice.
⇒ The reference case-study (section 3.4) has been chosen according to standards and results were compared with similar simulations. See section 3.4 “Consistency of the simulations has been verified. Firstly, the roundabout simulated has a typical size of 30m for the inner radius [30]. Secondly, in order to cover a wide range of operating range and to generate a tense traffic the simulation has been done with vehicles arriving randomly with a period of 2 seconds, which is an upper-bound to the vehicle capacity observed in this study.”
The purpose of section 4.1 is to demonstrate the flexibility of the simulation developed and the compatibility of the traffic manager with other case studies. The results were shown to check their consistency with the change of case study.
7) table 5.
- Please, explain how the combinations from 1 to 8 can be identified and understood by the reader.
- also explain what do -1 or 1 value stand for.
⇒ Done
8) Recently, there is an increasing interest in proactive road safety assessments based on the examination of the characteristics of the road to identify the presence of risk factors (i.e., https://doi.org/10.1016/j.aap.2022.106858 was carried out for roundabouts). How may your study be of support for such safety approaches? Can be the proactive approach potentially combined with your results? Please add a comment on the topic.
⇒ Added in section Conclusion : This study requires some installation and adaptation of the current roundabouts such as an external supervisor i.e. CSU which manages the traffic. Therefore, it is suggested to implement this strategy in roundabouts with a high risk factor as discussed by Rella Ricardi et al [35].
Round 2
Reviewer 2 Report (New Reviewer)
Thank you for addressing my comments.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The article discusses important aspects related to the introduction of autonomous vehicles and V2I communication. The authors indicated the structural aspects of the construction of roundabouts adapted to the movement of nuclear vehicles on them. The presented version of the article takes into account the comments made in the previous review.
Author Response
As indicated by the reviewer : The article discusses important aspects related to the introduction of autonomous vehicles and V2I communication. The authors indicated the structural aspects of the construction of roundabouts adapted to the movement of nuclear vehicles on them. The presented version of the article takes into account the comments made in the previous review.
==> so no new comments from our side
Reviewer 2 Report
The manuscript has been improved a lot according to the former suggestions. However, there are some formatting errors of the section headings, charts and references in the manuscript. It is recommended that the manuscript should be checked and modified carefully.
Author Response
The manuscript has been improved a lot according to the former suggestions. However, there are some formatting errors of the section headings, charts and references in the manuscript. It is recommended that the manuscript should be checked and modified carefully.
==> we added more references and we agree on we enhance the quality of some charts, we have also corrected some typo errors.
Reviewer 3 Report
Still Quality of the figures needs to Enhance
Author Response
Dear reviewer : we have improved the quality of all the figures, we have made sure that the legends and graphics are readable when printing the document or reading it on the screen.
Reviewer 4 Report
The subject matter of the paper is interesting, but after a thorough analysis of the content of the paper, the following issues were identified.
- how was the issue of "safety" in the title of the paper evaluated? There is no analysis of road incidents in the content, so the title should be changed;
- Abstract section: "The performance of the developed traffic manager is compared with SUMO." - what is compared to what?
- 1. Introduction section: there is no explanation what is the problem of driving an autonomous vehicle through the roundabout and not only through the roundabout, but generally through the network of roads and streets;
- 1.Intorduction section: no explanation of the concept of "ego vehicle";
- 1.Intorduction section: no explanation of "CSU" abbreviation; the explanation is only in section 2.1., i.e. the correct order of introduction and explanation of terms was not maintained;
- lack of a solid literature review and no separate "Literature review" section;
- there is no presentation of the current state of knowledge in the discussed problem of the passage of autonomous vehicles through the roundabout;
- the description of the vehicle model specified by "bicycle model" is missing; what bike model are you talking about?
- how the "moving block" model differs from other models used for a long time in traffic simulation, e.g. models: cells, following the leader, keeping a safe gap/headway?
- how was the correctness of the presented traffic control model verified? If only by simulation, how was the correctness of the simulation model verified?
- page 8: "the traffic congestion can be depicted in three phases" - incorrect use of the term "congestion"; because this term refers to the state of traffic saturation and then the state of overloading the roundabout with traffic, which results in an increase in the density of streams to the maximum value, which results in stopping the traffic and a traffic jam/congestion, i.e. congestion;
- Table 1 - there is no explanation for the presented parameters;
- formulas (1)-(7) - no explanation of variables and parameters; the explanations placed at the end of the paper in the table make it very difficult to analyze when reading the paper;
- section "4.4 CBVC comparison with Sumo" - what was compared? traffic simulation programs or the proposed traffic control model? this section does not imply that the developed traffic control model was compared to any other traffic control model or system; table 6 presents only the basic parameters of programs for vehicle traffic summation;
- very poor list of literature - only 20 items;
- there is no comparison of the proposed solution with other models related to the presented problem;
Author Response
Dear reviewer, first of all thank you very much for your comments, all constructive, we were able to further improve the quality of our paper by responding to your comments as indicated below
How was the issue of "safety" in the title of the paper evaluated? There is no analysis of road incidents in the content, so the title should be changed;
=> functional safety is implemented, meaning that the risk of vehicle while running in the roundabout is fixed thanks to an overall manager (CSU) which will manage all the vehicles in the roundabout. However, we agree that no study of the risks issues and the analysis hasn’t been done. Therefore, title has been changed.
- Abstract section: "The performance of the developed traffic manager is compared with SUMO." - what is compared to what?
=>Performance in terms of traffic are compared with existing algorithms.
- 1. Introduction section: there is no explanation what is the problem of driving an autonomous vehicle through the roundabout and not only through the roundabout, but generally through the network of roads and streets;
==> As mentioned in the introduction, the case studied in this paper is the roundabout as it is a difficult type of intersection to manage for CAV, that it is an important intersection in France, with a lot of roundabouts constructed each year.
- 1.Intorduction section: no explanation of the concept of "ego vehicle";
==> ego vehicle means the vehicle of interest. It has been replaced by ‘each vehicle’.
- 1.Intorduction section: no explanation of "CSU" abbreviation; the explanation is only in section 2.1., i.e. the correct order of introduction and explanation of terms was not maintained;
=> Corrected. We defined the CSU as the Central Signalling Unit in the abstract and in the introduction
- lack of a solid literature review and no separate "Literature review" section;
==> Introduction section is citing the different studies related to the studied topic.
- there is no presentation of the current state of knowledge in the discussed problem of the passage of autonomous vehicles through the roundabout;
==> Added some literature in the intro + comparison of the traffic manager with a benchmark car following type simulator
- the description of the vehicle model specified by "bicycle model" is missing; what bike model are you talking about?
==> Added a reference for the bicycle model
- how the "moving block" model differs from other models used for a long time in traffic simulation, e.g. models: cells, following the leader, keeping a safe gap/headway?
==> Precised in the text the difference between moving block architecture and car following architecture. Section 2.1
- how was the correctness of the presented traffic control model verified? If only by simulation, how was the correctness of the simulation model verified?
==> Firstly, the results obtained are comparable to other studies (average of 1400 veh/h) + Same order of magnitude as the results obtained with the benchmarked simulator , the case study is a typical roundabout size, the simulation has been done with vehicles arriving randomly in each entries each 2 seconds (Thus covering a wide range of cases) and if vehicles are colliding, the simulation stops and returns a flag. This flag hasn’t been seen for all the simulation runned
- page 8: "the traffic congestion can be depicted in three phases" - incorrect use of the term "congestion"; because this term refers to the state of traffic saturation and then the state of overloading the roundabout with traffic, which results in an increase in the density of streams to the maximum value, which results in stopping the traffic and a traffic jam/congestion, i.e. congestion;
==> Changed by accumulation phase and decumulation phase
- Table 1 - there is no explanation for the presented parameters;
Has been included in the text
- formulas (1)-(7) - no explanation of variables and parameters; the explanations placed at the end of the paper in the table make it very difficult to analyze when reading the paper;
==> Has been included in the text
- section "4.4 CBVC comparison with Sumo" - what was compared? traffic simulation programs or the proposed traffic control model? this section does not imply that the developed traffic control model was compared to any other traffic control model or system; table 6 presents only the basic parameters of programs for vehicle traffic summation.
==> Comparison in terms of traffic flow has been done with SUMO. This comparison was done to have a first intuition about the performance in terms of traffic flow of our developed CBVC. Performance of safety of the developed CVBC is inherent to its characteristics.
- very poor list of literature - only 20 items;
=> Added literature, comparison with existing studies on the case + reference to the bicycle model + comparison of other car following methods, now we do have 31 items.
- there is no comparison of the proposed solution with other models related to the presented problem;
==> Comparison with a benchmark simulator + with other studies (in terms of car following model, in terms of priorities (first arrived first served or in-roundabout priority), in terms of traffic flow, in terms of safety).
Author Response File: 
Author Response.pdf
Round 2
Reviewer 4 Report
There is still no explanation for the reader, i.e. in the content of the paper, of the following issues reported in the first review:
"- 1. Introduction section: there is no explanation what is the problem of driving an autonomous vehicle through the roundabout and not only through the roundabout, but generally through the network of roads and streets"
"- lack of separate "Literature review" section;"
"- the description of the vehicle model specified by "bicycle model" is missing;" Unfortunately, references to a source [23] with paid content and to a source unavailable [24] inadvertently exposes the reader to unnecessary costs; It is a disregard for the reader - the more so that the mentioned bicycle model is the first basic assumption of the discussed issues - page 3 of the manuscript;
"- how the "moving block" model differs from other models used for a long time in traffic simulation, e.g. models: cells, following the leader, keeping a safe gap/headway?"
"- how was the correctness of the presented traffic control model verified? If only by simulation, how was the correctness of the simulation model verified?"
"- section "4.4 CBVC comparison with Sumo" - what was compared? traffic simulation programs or the proposed traffic control model? this section does not imply that the developed traffic control model was compared to any other traffic control model or system; table 6 presents only the basic parameters of
programs for vehicle traffic summation."
"- there is no comparison of the proposed solution with other models related to the presented problem;"
