Integrating Autonomous Vehicles (AVs) into Urban Traffic: Simulating Driving and Signal Control
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe article analyzes the integration of autonomous vehicles (AVs) into urban traffic, focusing on signalized intersections and AV driving behaviors - cautious, normal, aggressive, and platooning - and their impact on key metrics such as queue lengths, travel times, vehicle delays, emissions, and fuel consumption. Using PTV VISSIM simulations, the study evaluates 21 scenarios with varying AV penetration rates (from 25% to 100%) at an intersection in Ankara, Turkey. The results show that aggressive and platooning AVs significantly improve traffic efficiency by reducing delays and emissions, while cautious behaviors lead to greater delays and fuel consumption. The study highlights the need for optimized signal control strategies and AV algorithms to maximize benefits in urban contexts.
The paper is well-structured and very clear, and the reviewer commends the authors for this.
However, the article is not particularly innovative. Previous studies have already examined both the effects of AV aggressiveness and the impact on isolated intersections. Moreover, the results are in line with many studies that demonstrate how more aggressive vehicles and platooning techniques can improve traffic flow.
Remarks on the content:
- In paragraph 3.4, starting from line 166, the desired speed of AVs is estimated using simulation data. However, simulation data in VISSIM is defined based on a desired speed set in advance. It is unclear how the authors calibrated a desired speed without predefining it in the software. Please explain the calibration method.
- The tables report data with precision levels that are not in line with the actual level of approximation of the measurements or simulations. Approximating speed to the third decimal point, as in Table 3, or queue lengths to the eighth decimal point is unrealistic and unnecessary for engineering problems. It is recommended to remove such levels of precision and reduce or eliminate Tables 3, 7, 8, 9, 10, and 11, presenting average values instead of all measurements.
- Paragraph 3.5 is overly detailed, given that it describes a model not developed by the authors. Are the values used the standard ones in VISSIM, or were they calibrated? If they were calibrated, please explain how. Alternatively, reduce and summarize the paragraph.
- The calibration of the scenario, described in paragraph 3.6, is insufficient. Paragraphs 3.9 and 3.10, which deal with model validation, should be integrated into the calibration section.
- The validation results show that the model simulates longer queues but shorter travel times. Please explain this apparent contradiction.
- Were the queue measurements at the intersection taken during the same signal cycle for all lanes?
- The figures illustrating the results are difficult to read due to closely packed lines with very similar colors. Consider modifying the representation using markers and different colors (e.g., a marker for each type of mix and a color for each level of aggressiveness distribution).
- How were the emissions calculated? Does it make sense to evaluate emissions considering that scenarios with 100% AVs are likely far in the future, and the vehicle fleet will probably be very different?
Author Response
|
Comments 1: In paragraph 3.4, starting from line 166, the desired speed of AVs is estimated using simulation data. However, simulation data in VISSIM is defined based on a desired speed set in advance. It is unclear how the authors calibrated a desired speed without predefining it in the software. Please explain the calibration method.
|
|
Response 1: Thank you for pointing this out. We agree with this comment. Therefore, we have addressed the calibration of desired speeds for AVs, we examined various speed scenarios, including speeds higher, lower, and equal to those of human-driven vehicles, to simulate different AV behaviors at varying penetration rates. The optimal speed was determined by observing how AVs interacted with human-driven vehicles under different traffic conditions, particularly in a congested environment. Although a desired speed was set in the simulation, the model naturally adjusted the actual speeds of AVs based on the surrounding traffic, considering factors like vehicle acceleration, deceleration, and lane-changing possibilities. This approach allowed us to identify the effective operating speed of AVs in real-world-like traffic scenarios. A clarification has been added to the manuscript to explain this matter. page 1&2, line 38-48. |
|
Comments 2: The tables report data with precision levels that are not in line with the actual level of approximation of the measurements or simulations. Approximating speed to the third decimal point, as in Table 3, or queue lengths to the eighth decimal point is unrealistic and unnecessary for engineering problems. It is recommended to remove such levels of precision and reduce or eliminate Tables 3, 7, 8, 9, 10, and 11, presenting average values instead of all measurements. |
|
Response 2: Agree. We have revised Tables 3, 7, 8, 9, 10, and 11 accordingly. The values have been rounded to 3 decimal digits to enhance clarity and simplicity, ensuring the precision is more aligned with the nature of the measurements and simulations.
Comments 3: Paragraph 3.5 is overly detailed, given that it describes a model not developed by the authors. Are the values used the standard ones in VISSIM, or were they calibrated? If they were calibrated, please explain how. Alternatively, reduce and summarize the paragraph.
Response 3: Thank you for your comment. The car-following and lane-changing sections are central to demonstrating the differences between AVs and human-driven vehicles. We aimed to provide a thorough explanation of these behaviors, supported by relevant literature, to highlight the significance of these models. The parameters used in our simulations are the standard values from the VISSIM model manual, which have been widely adopted in other studies. Therefore, we felt it was important to maintain the detail in these sections to ensure clarity on these critical aspects.
Comments 4: The calibration of the scenario, described in paragraph 3.6, is insufficient. Paragraphs 3.9 and 3.10, which deal with model validation, should be integrated into the calibration section.
Response 4: Agree. We have integrated sections 3.9 and 3.10, which address model validation, into the calibration section. The updated section is now titled 3.8 Model Calibrations and Validation, with subsections 3.8.1 Average Queue Length Validation and 3.8.2 Average Travel Time Validation. This integration provides a more cohesive discussion of the calibration and validation process in our study. pages 16-22
Comments 5: The validation results show that the model simulates longer queues but shorter travel times. Please explain this apparent contradiction.
Response 5: Agree. The contradiction between longer queues and shorter travel times can be explained by differences between real-world human driving and the simulated model. In the simulation, longer queues may form due to conservative driving behaviors, with larger gaps or slower reactions at intersections. However, once vehicles start moving, they tend to accelerate more efficiently in the simulation, leading to smoother traffic flow and shorter travel times. This explains why, despite longer queues, vehicles clear the intersection faster in the model compared to real-world conditions. page 21, line 446-452.
Comments 6: Were the queue measurements at the intersection taken during the same signal cycle for all lanes?
Response 6: Agree. We have revised the explanation to clarify that the queue length measurements were taken during the red signal phase across all lanes simultaneously for each direction, at different signal cycle times. page 18, line 401-406.
Comments 7: The figures illustrating the results are difficult to read due to closely packed lines with very similar colors. Consider modifying the representation using markers and different colors (e.g., a marker for each type of mix and a color for each level of aggressiveness distribution).
Response 7: Agree. Thank you for your valuable suggestion. We have revised the figures based on your feedback, using distinct markers for each type of mix and different colors to represent each level of aggressiveness distribution. Additionally, we have replaced the data point results with regression trend lines based on the third reviewer's feedback to enhance clarity and make the figures easier to interpret. pages 23-28
Comments 8: How were the emissions calculated? Does it make sense to evaluate emissions considering that scenarios with 100% AVs are likely far in the future, and the vehicle fleet will probably be very different?
Response 8: Agree. We have added Section 3.9: Emission Modeling in VISSIM to explain the calculation of emissions using HBEFA. Additionally, the relevance of evaluating emissions under a 100% AV scenario is addressed in Section 5.2: Environmental Impacts: Emissions and Fuel Consumption, highlighting the study's role in informing long-term transportation planning despite future changes in vehicle fleets. page 22, 29, line 456-486, 659-666.
|
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThis manuscript presents a simulation study regarding the integration of autonomous vehicles in urban traffic systems. The authors evaluate different driving behaviors and demonstrate the potential of autonomous vehicles to enhance traffic efficiency. I believe that the authors have invested significant effort into this work. However, there are some aspects that require further attention. I encourage the authors to consider my comments below.
1. There is a lack of detail on vehicle platoons in the introduction. Explain the benefits of vehicle platoon and cite relevant works, such as https://doi.org/10.1109/ACCESS.2024.3428341
2. Table 1 is difficult to follow. Please consider commenting the difference between literature works along the text.
3. Figure quality needs to be improved. Check Figures 2, 6, 13, 14, 15, 16, 17, 18, 19.
4. Modify Table 2 so that it does not split between pages.
5. Table 3 is difficult to follow. Include the vehicle speed either in m/s or km/h, but not in both units, as it does not provide additional details.
6. Provide an adequate literature source for the equation (2).
7. It is recommended that the discussion and conclusion sections include some numerical indicators that demonstrate the relevance of the presented work.
8. In line 506, the authors state “However, the aggressive AV behavior in our simulations also introduced instability at higher signal cycle times, leading to increased vehicle oscillations and traffic disturbances”. It would be greatly beneficial if some solutions could be provided to this issue, as instability in traffic is a common problem.
9. In line 509, the authors state “The results demonstrated that platooning AVs can significantly reduce emissions (CO, NOx, VOC) and fuel consumption, particularly at shorter cycle times”. It is necessary to determine whether this reduction is significant enough to justify the cost of implementing this new technology.
10. It is recommended that a list of abbreviations be included at the end of the manuscript, in accordance with the templates provided by MDPI.
Author Response
|
Comments 1: There is a lack of detail on vehicle platoons in the introduction. Explain the benefits of vehicle platoons and cite relevant works
|
Response 1: Agree. We have carefully addressed your suggestion to expand on the concept of vehicle platooning in the introduction. The revised introduction now explains the benefits of vehicle platooning, including improved traffic flow, fuel efficiency, and safety, as well as its impact on reducing gaps between vehicles and enhancing intersection performance. We have also included the suggested reference [https://doi.org/10.1109/ACCESS.2024.3428341]. page 1&2, line 38-48.
Comments 2: Table 1 is difficult to follow. Please consider commenting the difference between literature works along the text
Response 2: Agree. We have added commentary in the Literature Review section to highlight the differences between the studies in Table 1. These clarifications explain key insights and how they relate to our research. page 3, line 121-147.
Comments 3: Figure quality needs to be improved. Check Figures 2, 6, 13, 14, 15, 16, 17, 18, 19.
Response 3: Agree. We have improved the resolution of Figures 2, 6, 13, 14, 15, 16, 17, 18, and 19.
Comments 4: Modify Table 2 so that it does not split between pages.
Response 4: Thank you, we have modified the Table 2 so that it does not split between pages.
Comments 5: Table 3 is difficult to follow. Include the vehicle speed either in m/s or km/h, but not in both units, as it does not provide additional details.
Response 5: Agree. We have revised Table 3 by removing the vehicle speed in m/s and retaining only the speed in km/h for clarity and consistency.
Comments 6: Provide an adequate literature source for the equation (2).
Response 6: Agree. We have provided and cited the appropriate reference for Equation (2). The reference is now cited as [44]: Treiber, M.; Hennecke, A.; Helbing, D. Congested Traffic States in Empirical Observations and Microscopic Simulations. Phys. Rev. E 2000, 62, 1805, doi:10.1103/PhysRevE.62.1805. page 12, line 287
Comments 7: It is recommended that the discussion and conclusion sections include some numerical indicators that demonstrate the relevance of the presented work.
Response 7: Agree. We have revised the manuscript to include specific numerical results in both the discussion and conclusion sections, highlighting the impacts of various AV behaviors on traffic metrics. page 28-31
Comments 8: In line 506, the authors state “However, the aggressive AV behavior in our simulations also introduced instability at higher signal cycle times, leading to increased vehicle oscillations and traffic disturbances”. It would be greatly beneficial if some solutions could be provided to this issue, as instability in traffic is a common problem.
Response 8: Agree. We have addressed the issue regarding the instability caused by aggressive AV behavior at higher signal cycle times by providing potential solutions in the revised manuscript. These include adaptive traffic signal control systems and cooperative AV algorithms to reduce vehicle oscillations and improve traffic stability. page 28&29, line 612-621.
Comments 9: In line 509, the authors state “The results demonstrated that platooning AVs can significantly reduce emissions (CO, NOx, VOC) and fuel consumption, particularly at shorter cycle times”. It is necessary to determine whether this reduction is significant enough to justify the cost of implementing this new technology.
Response 9: Agree. We have addressed your comment by discussing the cost-effectiveness of platooning AV technology, noting that emission and fuel reductions can justify the cost over time. This is supported by references in the revised manuscript. page 29, line 646-652.
Comments 10: It is recommended that a list of abbreviations be included at the end of the manuscript, in accordance with the templates provided by MDPI.
Response 10: Agree. We have included a "List of Abbreviations" at the end of the manuscript. page 34.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe article presents a timely investigation into the impact of autonomous vehicles on urban traffic flow. The authors focus on a signalized intersection, a common bottleneck in urban environments. The study employs a microscopic simulation approach using PTV VISSIM software, which allows for detailed modeling of individual vehicle behavior. The authors consider a range of AV driving behaviors, including cautious, normal, aggressive, and platooning, to assess their effects on traffic metrics such as queue lengths, travel times, and emissions. The research provides valuable insights into the potential benefits and challenges of integrating AVs into existing urban traffic systems. However, the paper would benefit from a more comprehensive discussion of the limitations of the simulation model and the generalizability of the findings to real-world settings.
Overall, the study offers a solid foundation for understanding the complexities of AV integration at signalized intersections. Further research could explore the impact of different signal control strategies and the interaction between AVs and other road users, such as pedestrians and cyclists.
Questions:
1. Given the significant variability in driving conditions and driver behavior in real-world urban environments, how might the authors address the limitations of relying solely on simulation-based results to draw conclusions about the impact of AVs on traffic flow?
2. The study focuses on a single signalized intersection. How could the research be expanded to consider the broader network-level effects of AV integration, including potential impacts on traffic flow on adjacent roadways and intersections?
Further issues to be solved/fixed:
1. Please remove the (ugly) black, thick frames from all the figures.
2. Please prepare a "List of abbreviations" and a "Nomenclature" (i.e., "List of symbols") at the end of the manuscript.
3. Please prove/certify that you have a min. research licence of PTV VISSIM
4. Please supplement the citation/references (sources) from Fig. 1 and similar items.
5. Please use the "Palatino Linotype" font type in all the objects (figures, tables, etc.) (e.g., see Fig. 2, etc.)
6. Please supplement the units in all the tables and diagrams (e.g., see Table 2)
7. Please reformat/redraw Fig. 3 because there are a lot of labels that cannot be read entirely.
8. Please add units to the symbols where you explain their meanings at their first appearance! Please indicate if they are unitless parameters!
9. Please use "–" (Alt+0150) for negative sign (instead of "-") in the entire manuscript, even in all the objects!
10. Table 4: "m/s2", please put "2" in the upper case everywhere. Please make a double-check in the whole submission!
11. Table 7: please do not use too many decimal digits in the values. Please round them to 3 decimal digits, it may be enough for the comparison(s) and evaluation(s). Please make a double-check in the whole submission!
12. Fig. 11: please add values/labels for the columns. Please make a double-check in the whole submission!
13. Fig. 13-19: can you add trend lines or regression lines to the simulated results (results from the simulations)?
14. Figs. 13-19: can you extrapolate assumed results from the extrapolated values?
15. see comments #13-#14: can you formulate models for them? Can you calibrate and validate them?
16. The Conclusions must be rewritten, focusing on the obtained results with values! The half of the Conclusions is about the future research possibilities! Please improve the ratio/balance!
Author Response
|
Comments 1: Please remove the (ugly) black, thick frames from all the figures.
|
Response 1: Agree. The black, thick frames from all the figures have been deleted as requested.
Comments 2: Please prepare a "List of abbreviations" and a "Nomenclature" (i.e., "List of symbols") at the end of the manuscript.
Response 2: Agree. We have included a "List of Abbreviations" at the end of the manuscript. page 34.
Comments 3: Please prove/certify that you have a min. research licence of PTV VISSIM.
Response 3: We have mentioned in the Acknowledgment section that the research was conducted using a minimum research license of PTV VISSIM, as required. page 32. Line 781-782
Comments 4: Please supplement the citation/references (sources) from Fig. 1 and similar items.
Response 4: Agree. We have mentioned the source of each figure, including Fig. 1 and similar items, in the corresponding captions and citation sections as requested.
Comments 5: Please use the "Palatino Linotype" font type in all the objects (figures, tables, etc.) (e.g., see Fig. 2, etc.).
Response 5: Agree. We have considered your request and reformatted the font type in all objects, including figures and tables, to "Palatino Linotype" as required.
Comments 6: Please supplement the units in all the tables and diagrams (e.g., see Table 2).
Response 6: Agree. We have supplemented the units in all the tables and diagrams, including Table 2, as requested.
Comments 7: Please reformat/redraw Fig. 3 because there are a lot of labels that cannot be read entirely.
Response 7: Agree. We have reformatted and redrawn Fig. 3 to ensure that all labels are clearly readable, as requested.
Comments 8: Please add units to the symbols where you explain their meanings at their first appearance! Please indicate if they are unitless parameters!
Response 8: Agree. We have added units to the symbols where their meanings are explained at their first appearance in the document. For parameters that are unitless, this has been clearly indicated.
Comments 9: Please use "–" (Alt+0150) for negative sign (instead of "-") in the entire manuscript, even in all the objects!
Response 9: Agree. We have replaced the negative sign ("-") with the – dash ("–", Alt+0150) throughout the entire manuscript, including in all objects such as figures, tables, and equations, as requested.
Comments 10: Table 4: "m/s2", please put "2" in the upper case everywhere. Please make a double-check in the whole submission!
Response 10: Agree. We have adjusted the formatting of "m/s²" to ensure the "²" is in superscript throughout Table 4 and the entire submission.
Comments 11: Table 7: please do not use too many decimal digits in the values. Please round them to 3 decimal digits, it may be enough for the comparison(s) and evaluation(s). Please make a double-check in the whole submission!
Response 11: Agree. We have rounded the values in Table 7 to 3 decimal digits for clarity and simplicity, as it is sufficient for comparisons and evaluations. A complete double-check has also been conducted throughout the entire submission to ensure consistency in rounding across all tables and data points.
Comments 12: Fig. 11: please add values/labels for the columns. Please make a double-check in the whole submission!
Response 12: Agree. We have added values and labels to the columns in Fig. 11 & 12 as requested.
Comments 13: Fig. 13-19: can you add trend lines or regression lines to the simulated results (results from the simulations)?
Response 13: Agree. We have added the regression lines to the simulated results, but the graph became quite complicated and difficult to interpret with all the additional information. Therefore, we decided to keep only the regression lines for clarity and to enhance understanding of the trends in the data.
Comments 14&15: Figs. 13-19: can you extrapolate assumed results from the extrapolated values? see comments #13-#14: can you formulate models for them? Can you calibrate and validate them?
Response 14&15: While we appreciate the importance of extrapolating assumed results and generating models for the data presented in Figures 13-19 (including queue lengths, travel times, delays, CO, NOx, VOC, and fuel consumption), the complexity and extensive analysis required for this process exceeds the scope of our current work.
We believe this would make an excellent suggestion for future studies, where a more in-depth exploration can be conducted to calibrate and validate these models effectively. Thank you for your understanding!
Comments 16: The Conclusions must be rewritten, focusing on the obtained results with values! The half of the Conclusions is about the future research possibilities! Please improve the ratio/balance!
Response 16: Agree. We have rewritten the Conclusions to place greater emphasis on the obtained results, incorporating specific values to highlight key findings. While we recognize the importance of discussing future research possibilities, we have improved the balance by reducing that section to ensure a more focused presentation of the results. Thank you for your feedback!
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have made substantial improvements to the manuscript, which I believe is now suitable for publication. I would like to extend my congratulations on a job well done.
Author Response
Response 1: Thank you very much for your kind words and positive feedback. We are grateful for your thorough review and valuable suggestions, which have significantly improved the quality of our manuscript. Your insights have been instrumental in shaping the final version, and we are delighted to hear that you now consider it suitable for publication. We appreciate the time and effort you dedicated to reviewing our work.
|
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsDear Authors,
many thanks for your valuable revisions.
There remained issues to be fixed:
1. Please modify the title to the following simplified one: "Integrating Autonomous Vehicles (AVs) into Urban Traffic: Simulating Driving and Signal Control"
2. Please try to eliminate using abbreviations in the Abstract (e.g., EV, etc.). In the keywords you can leave/keep them, of course!
3. The numbering of the references in the text is not appropriate, e.g., see rows #37 and #43, etc. The order is harmed. Please renumber considering the first appearance in the main text!
4. At the end of Section 2, you must write min. a paragraph summarizing and criticizing the collected literature and point out the research gap of your study and how to solve this problem.
5. In Table 2, please use two decimal digits for all the values, i.e., for "24.00", "25.00" and "21.60", too. Please fix them from "24", "25" and "21.6".
6. Figs. 4 and 5: please add units for the vertical axes! (e.g., "[-]").
7. Fig. 6: please use italic style for "deltav" in the horizontal axis!
8. Table 6: please see the comment #5 (two decimal digits are required for all the values!)
9. List of Symbols (Nomenclature) is missing! The Reviewer requested it in the previous review as compulsory item! Please supplement it!
Yours sincerely,
The Reviewer
Author Response
|
Comments 1: Please modify the title to the following simplified one: "Integrating Autonomous Vehicles (AVs) into Urban Traffic: Simulating Driving and Signal Control".
|
Response 1: Agree. We have revised the title as per your recommendation to "Integrating Autonomous Vehicles (AVs) into Urban Traffic: Simulating Driving and Signal Control" to make it more concise and reflective of the study's focus, thank you.
Comments 2: Please try to eliminate using abbreviations in the Abstract (e.g., EV, etc.). In the keywords you can leave/keep them, of course!
Response 2: Agree. We have revised the abstract to eliminate the use of abbreviations such as "EV" and ensured that the full terms are used for clarity.
Comments 3: The numbering of the references in the text is not appropriate, e.g., see rows #37 and #43, etc. The order is harmed. Please renumber considering the first appearance in the main text!.
Response 3: Agree. Thank you for pointing this out. We have carefully reviewed the numbering of the references in the text and corrected the order to ensure that they now follow the sequence of their first appearance in the main text.
Comments 4: At the end of Section 2, you must write min. a paragraph summarizing and criticizing the collected literature and point out the research gap of your study and how to solve this problem.
Response 4: Agree. We have written a paragraph summarizing the importance of our work. page 6, line 160-169.
Comments 5: In Table 3, please use two decimal digits for all the values, i.e., for "24.00", "25.00" and "21.60", too. Please fix them from "24", "25" and "21.6".
Response 5: Agree. We have revised Table 3 to ensure that all values are presented with two decimal digits, correcting "24," "25," and "21.6" to "24.00," "25.00," and "21.60," respectively. page 9&10.
Comments 6: Figs. 4 and 5: please add units for the vertical axes! (e.g., "[-]").
Response 6: Agree. Thank you we have updated Figs. 4 and 5. page 10&11.
Comments 7: Fig. 6: please use italic style for "delta v" in the horizontal axis!
Response 7: Agree. We have revised Fig. 6 to display "delta v" in italics on the horizontal axis, as requested. page 11.
Comments 8: Table 6: please see the comment #5 (two decimal digits are required for all the values!)
Response 8: Agree. We have revised Table 6 to ensure that all values are presented with two decimal digits.
Comments 9: List of Symbols (Nomenclature) is missing! The Reviewer requested it in the previous review as compulsory item! Please supplement it!
Response 9: Dear reviewer kindly, as I checked the template of Applied Science and some of MDPI journals I couldn’t find a specific details or place in the manuscript to insert the abbreviations - Symbols therefore we have included the list of Abbreviations at the end of the manuscript. page 36.
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
Comments and Suggestions for AuthorsEverything is fine, there is only one exception: the List of Symbols (Nomenclature) is missing. The reviewer requested its preparation and incorporation two times, but it has not been done already. Please supplement it!
Author Response
|
Comments 1: Everything is fine, there is only one exception: the List of Symbols (Nomenclature) is missing. The reviewer requested its preparation and incorporation two times, but it has not been done already. Please supplement it!
|
Response 1: Dear Reviewer,
Thank you for your thorough feedback and suggestions. I would like to inform you that both the List of Abbreviations (line 791) and the List of Symbols (Nomenclature) (line 807) have now been included in the manuscript. I apologize for any oversight in the previous versions and appreciate your patience.
Please let me know if any further adjustments are needed.
