Review Reports - A Reproducible Evaluation Method for Intelligent-Driving Longitudinal Control Under Complex Weather Through Operational Design Domain Parameter Perturbation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The abbreviations used in the abstract should be mentioned within the bracket, whenever it appears first. E.g. AEB, TTC, SCANeR. When abbreviations are not properly introduced, it can negatively impact the reader’s experience.

The method considered meteorological factors such as snow, fog and rain. Why Crosswinds / Strong Gusts that affect the longitudinal stability of the vehicle are not included ? can these be parameterised into graded level to have a comprehensive evaluation method ?

Post braking gaps reduced at higher vehicle speed, compared to low speed in all weather conditions. It is suggested that an explicit description is required to technically present the complex behaviour of the vehicle.

Refer to Page no. 17, second paragraph. “Under the same initial speed and relative-motion conditions, this delay passively reduces the available braking distance, thereby amplifying the risk of rear-end collision”. How the front-end collision is addressed ?

Detailed explanation is necessary about questionnaire-based approach to quantify preceived risk. Was it structured, semi-structured, descriptive ? A sample questionnaire is helpful to understand more about the survey.

The explanation of risk and functional reliability of sensor is currently limited in the current version of the manuscript; expanding it with literature-based insights and a more thorough discussion would improve the quality of the manuscript. The following risk and reliability litereture may be helpful if added as references.
Subramaniam, S., Faisal, A. I., & Deen, M. J. (2022). Wearable sensor systems for fall risk assessment: A review. Frontiers in digital health, 4, 921506.
Yan, C., Xu, W., & Liu, J. (2016). Can you trust autonomous vehicles: Contactless attacks against sensors of self-driving vehicle. Def Con, 24(8), 109.
Loganathan, M. K., Goswami, P., & Bhagawati, B. (2016). Failure evaluation and analysis of mechatronics-based production systems during design stage using structural modeling. Applied Mechanics and Materials, 852, 799-805.
Habchi, G., & Barthod, C. (2016). An overall methodology for reliability prediction of mechatronic systems design with industrial application. Reliability Engineering & System Safety, 155, 236-254.
Askari, H., Khajepour, A., Khamesee, M. B., & Wang, Z. L. (2019). Embedded self-powered sensing systems for smart vehicles and intelligent transportation. Nano Energy, 66, 104103.
Loganathan, M. K., & Gandhi, O. P. (2015). Reliability evaluation and analysis of CNC cam shaft grinding machine. Journal of Engineering, Design and Technology, 13(1), 37-73.

Author Response

（1）The abbreviations used in the abstract should be mentioned within the bracket, whenever it appears first. E.g. AEB, TTC, SCANeR. When abbreviations are not properly introduced, it can negatively impact the reader’s experience.

Respond: Thank you for this helpful comment. We addressed this point by substantially rewriting the abstract and minimizing abbreviations in the abstract altogether, so that the abstract is more self-contained and easier to read. The revised abstract now presents the background, method, validation framework, main findings, and engineering significance more clearly without relying on unexplained abbreviations (Lines 17–32).

（2）The method considered meteorological factors such as snow, fog and rain. Why Crosswinds / Strong Gusts that affect the longitudinal stability of the vehicle are not included ? can these be parameterised into graded level to have a comprehensive evaluation method ?

Respond: Thank you for this insightful comment. We clarified the study scope in the revised manuscript. The present work focuses on longitudinal collision-mitigation performance in a lead-vehicle sudden-braking scenario. Accordingly, the selected perturbation factors are visibility-related weather degradation and road-adhesion reduction, because these two classes of factors directly affect the perception–decision–braking chain and the stopping safety margin of longitudinal control. We also explicitly state that crosswinds and strong gusts are more strongly coupled with lateral stability, yaw dynamics, and lane-keeping behavior, and are therefore left for future extension of the framework (Lines 106–125).

（3）Post braking gaps reduced at higher vehicle speed, compared to low speed in all weather conditions. It is suggested that an explicit description is required to technically present the complex behaviour of the vehicle.

Respond: Thank you. We strengthened the technical interpretation accordingly. In the revised manuscript, we now explain that the stronger reduction in post-braking gap at higher initial speed results from the combined effects of increased kinetic-energy dissipation demand, limited longitudinal tire force under low adhesion, and shortened effective braking distance in the presence of trigger delay (Lines 557–566).

（4）Refer to Page no. 17, second paragraph. “Under the same initial speed and relative-motion conditions, this delay passively reduces the available braking distance, thereby amplifying the risk of rear-end collision”. How the front-end collision is addressed ?

Respond: Thank you for pointing out this ambiguity. We clarified the collision interpretation in both the Results and the Discussion. The revised manuscript now explicitly refers to a frontal impact of the ego vehicle with the rear of the lead vehicle, i.e., a rear-end collision scenario (Lines 606–610; Lines 721–726).

（5）Detailed explanation is necessary about questionnaire-based approach to quantify preceived risk. Was it structured, semi-structured, descriptive ? A sample questionnaire is helpful to understand more about the survey.

Respond: Thank you for this important comment. We agree that the previous wording was insufficiently transparent. In the revised manuscript, rather than retaining an unsupported questionnaire-based claim, we revised Section 2.2 to describe this step more accurately as a preliminary expert-elicitation-based risk-grading mechanism for weather–road coupled states. We further clarified that this step is used as an engineering prioritization tool for test planning and boundary-oriented evaluation, rather than as a standalone statistical validation procedure (Lines 233–264).

（6）The explanation of risk and functional reliability of sensor is currently limited in the current version of the manuscript; expanding it with literature-based insights and a more thorough discussion would improve the quality of the manuscript. The following risk and reliability litereture may be helpful if added as references.

Subramaniam, S., Faisal, A. I., & Deen, M. J. (2022). Wearable sensor systems for fall risk assessment: A review. Frontiers in digital health, 4, 921506.
Yan, C., Xu, W., & Liu, J. (2016). Can you trust autonomous vehicles: Contactless attacks against sensors of self-driving vehicle. Def Con, 24(8), 109.
Loganathan, M. K., Goswami, P., & Bhagawati, B. (2016). Failure evaluation and analysis of mechatronics-based production systems during design stage using structural modeling. Applied Mechanics and Materials, 852, 799-805.
Habchi, G., & Barthod, C. (2016). An overall methodology for reliability prediction of mechatronic systems design with industrial application. Reliability Engineering & System Safety, 155, 236-254.
Askari, H., Khajepour, A., Khamesee, M. B., & Wang, Z. L. (2019). Embedded self-powered sensing systems for smart vehicles and intelligent transportation. Nano Energy, 66, 104103.
Loganathan, M. K., & Gandhi, O. P. (2015). Reliability evaluation and analysis of CNC cam shaft grinding machine. Journal of Engineering, Design and Technology, 13(1), 37-73.

Respond: Thank you for this valuable suggestion. We agree that the previous version did not discuss sufficiently the system-level risk implications of adverse weather and the functional reliability of the perception–decision–control chain. In the revised manuscript, we expanded the literature review and discussion accordingly. Specifically, we strengthened the Introduction by adding literature related to sensor vulnerability, mechatronic-system reliability, and intelligent-vehicle sensing (Lines 52–79), and we further expanded the Discussion to explain how precipitation, fog, and surface contamination may degrade the effective sensing quality of cameras, LiDAR, and radar through scattering, attenuation, occlusion, and signal distortion, thereby increasing target-detection uncertainty and trigger delay (Lines 699–714).

In response to the Reviewer’s recommendation, we carefully reviewed the suggested literature and incorporated the references most relevant to the present study—particularly those concerning autonomous-vehicle sensor vulnerability, mechatronic-system reliability, and intelligent-transportation sensing—into the revised manuscript and reference list (References 29–32, Lines 900–908).

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper focuses on the degradation of intelligent driving longitudinal control performance under complex weather conditions and proposes an evaluation method that combines ODD parameter perturbation with fuzzy comprehensive evaluation. The method is validated using the SCANeR platform and a digital twin system. The research topic is of practical relevance, particularly in the context of safety validation and ODD boundary identification for autonomous driving systems, and it demonstrates certain engineering application value. The overall structure of the paper is relatively complete, the methodological framework is reasonably systematic, and the experimental design is generally sound. However, from the perspectives of academic rigor and methodological innovation. It is recommend ed for acceptance after minor revision. The question are as following.

The manuscript only mentions that a “questionnaire-based method” was adopted, without providing details on the questionnaire design, sample size, or statistical analysis. This weakens the interpretability of the risk grading results, making it difficult for readers to assess their validity. As a result, the grading lacks reproducibility and credibility. It is recommended to supplement information on the number of participants, their professional background, the scoring method, and to include basic statistical analysis.
The key information regarding the expert consultation process—such as the number of experts and the distribution of their qualifications—is not disclosed. Furthermore, the evaluation matrix R (i.e., the membership degrees of each indicator to “excellent/good/fair/poor/very poor”) is entirely derived from subjective expert judgment. This purely experience-based approach may result in significant variability depending on individual experts and contradicts the claimed “data-driven objectivity” of the study. It is recommended to define explicit membership functions, validate the rationality of the membership assignments, and conduct sensitivity analysis.
Figures 6–8 do not include error bars or statistical significance indicators, and Figure 9 lacks quantitative comparative analysis. In addition, some sentences are overly long and should be simplified for better clarity.

Comments on the Quality of English Language

Some terms referring to the same indicators should be used consistently throughout the manuscript. Due to pagination, the readability of certain figures is reduced and could be improved through better layout adjustments.

Author Response

（1）The manuscript only mentions that a “questionnaire-based method” was adopted, without providing details on the questionnaire design, sample size, or statistical analysis. This weakens the interpretability of the risk grading results, making it difficult for readers to assess their validity. As a result, the grading lacks reproducibility and credibility. It is recommended to supplement information on the number of participants, their professional background, the scoring method, and to include basic statistical analysis.

Respond: Thank you. We agree that the previous description was insufficiently transparent. In the revised manuscript, we removed the unsupported “questionnaire-based method” wording and revised Section 2.2 to clarify that the risk-grading step is used as a preliminary expert-elicitation-based engineering prioritization tool for distinguishing low-, medium-, and high-risk coupled states before the subsequent metric-based and comprehensive evaluation stages (Lines 233–264).

（2）The key information regarding the expert consultation process—such as the number of experts and the distribution of their qualifications—is not disclosed. Furthermore, the evaluation matrix R (i.e., the membership degrees of each indicator to “excellent/good/fair/poor/very poor”) is entirely derived from subjective expert judgment. This purely experience-based approach may result in significant variability depending on individual experts and contradicts the claimed “data-driven objectivity” of the study. It is recommended to define explicit membership functions, validate the rationality of the membership assignments, and conduct sensitivity analysis.

Respond: Thank you for this important comment. In the revised manuscript, we clarified the methodological role of the preliminary risk-grading step and revised Section 2.3 to distinguish more explicitly between the objective component introduced at the weighting stage and the expert-dependent settings involved in fuzzy aggregation. We also acknowledge more explicitly that some evaluation settings still involve expert-dependent elements and identify their further refinement under stronger data constraints as an important future direction (Lines 266–366; Lines 797–810).

（3）Figures 6–8 do not include error bars or statistical significance indicators, and Figure 9 lacks quantitative comparative analysis. In addition, some sentences are overly long and should be simplified for better clarity.

Respond: Thank you. Rather than adding unsupported significance claims, we revised the wording of the Results section to frame the present validation more appropriately as a controlled comparative analysis across weather–road states (Lines 528–532; Lines 585–590). We also strengthened the quantitative interpretation associated with the comprehensive evaluation results and the differentiated degradation patterns under rain, fog, and snow (Lines 658–692). In addition, long and overly compact sentences were simplified throughout the manuscript to improve readability.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

First, the literature review remains insufficiently comprehensive and somewhat fragmented. While key standards, like SAE, ISO and recent perception studies are cited, the paper lacks deeper engagement with vehicle dynamics and tyre–road interaction research under varying load and environmental conditions (for example doi: 10.7307/ptt.v36i1.265) which provides valuable insights into tyre behavior, load sensitivity, and safety implications. This would significantly enhance the physical grounding of the friction-coefficient assumptions and strengthen the interpretation of braking performance degradation.

Second, the methodological transparency requires improvement. The selection of membership functions, expert weighting process, and questionnaire-based risk grading lack sufficient detail for reproducibility. For example, the criteria for expert selection, sample size justification, and sensitivity of results to weighting choices are not clearly discussed. This raises concerns about the robustness and transferability of the evaluation framework.

Third, although the co-simulation platform is sophisticated, the validation remains limited to simulated environments. The absence of any real-world or experimental benchmarking reduces confidence in the practical applicability of the findings. At minimum, the authors should discuss calibration strategies or provide comparative references to empirical data.

Finally, the discussion could be deepened by more explicitly linking results to control-system design implications, AEB tuning, adaptive thresholds under low adhesion. Additionally, uncertainty analysis and limitations of the TTC-based model should be elaborated.

Author Response

Respond: Thank you for this thoughtful and constructive comment. In response, we revised the manuscript in four aspects. First, we strengthened the literature review and physical grounding of the study by adding discussion of tyre–road interaction, braking-related dynamics, and system-level sensing/actuation/control reliability, thereby improving the interpretation of friction-coefficient assumptions and braking-performance degradation under adverse-weather conditions (Lines 52–79; 154–166; 204–209). Second, we improved methodological transparency by revising Section 2.2 to clarify that the risk-grading step is used as a preliminary expert-elicitation-based engineering prioritization tool rather than a standalone statistical validation procedure, and by revising Section 2.3 to explain more clearly the purpose of the hierarchical indicator system, the rationale of the combined-weight method, and the role of fuzzy evaluation in representing gradual degradation under coupled weather–road conditions (Lines 233–366). Third, we addressed the concern regarding the limitation of simulation-only validation by explicitly stating that the present findings are based on co-simulation and digital-twin validation rather than direct real-vehicle experiments, and that extrapolation to real-vehicle performance requires subsequent calibration and back-to-back validation (Lines 757–766; 797–809). Finally, we deepened the Discussion by linking the observed results more explicitly to control-design implications, including the need to jointly consider temporal response degradation and spatial safety-margin reduction under adverse weather, and by clarifying that the TTC-based controller is a simplified comparative control structure rather than a full production-grade AEB system; we also elaborated the limitations and future directions concerning uncertainty, transferability, and tighter calibration against real-vehicle or hardware-in-the-loop validation (Lines 421–450; 715–756; 797–810).

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This manuscript presents an evaluation of emergency braking under complex weather scenarios. Overall, the manuscript is challenging to follow in its current state. It is not clear what authors did in comparison to previous works. Nevertheless, I consider the work relevant enough for publication, so I encourage the authors to improve the manuscript. Please find my suggestions below:

The introduction should discuss the relevance of robust control methodologies for ensuring safe longitudinal vehicle motion. This is particularly relevant for vehicle platooning applications, which might be worth noting. Authors should cite and discuss recent, related works, such as Robust adaptive control of heterogeneous vehicle platoons in the presence of network disconnections with a novel string stability guarantee, Static output feedback control for vehicle platoons with robustness to mass uncertainty.
Please cite any source that can validate tables 1 and 2.
The weather state in Table 4 is difficult to understand, please provide more details on that.
Table 5 must be improved.
I don't see the purpose of section 2.3.2, please revise and provide more physical details on that.
Please provide more insights about SCANeR Studio software.
More details about the AEB system used in this work are required. Simulation results depend on the AEB structure, so this is necessary for reproducibility.
Maybe the TTC-based AEB should be presented as part of the methodology, and not straight into the simulation section.

Author Response

（1）The introduction should discuss the relevance of robust control methodologies for ensuring safe longitudinal vehicle motion. This is particularly relevant for vehicle platooning applications, which might be worth noting. Authors should cite and discuss recent, related works, such as Robust adaptive control of heterogeneous vehicle platoons in the presence of network disconnections with a novel string stability guarantee, Static output feedback control for vehicle platoons with robustness to mass uncertainty.

Respond: Thank you for this helpful suggestion. We revised the Introduction accordingly. The revised manuscript now relates the present study more explicitly to the broader context of robust longitudinal control under uncertainty, including car-following and platooning-related problems, and cites recent related literature (Lines 67–79; References 26–27, Lines 893–897).

（2）Please cite any source that can validate tables 1 and 2.

Respond: Thank you. We revised the manuscript to provide clearer source support for Tables 1 and 2. Table 1 is now explicitly linked to the meteorological industry standard Weather Risk Warning Levels for Expressway Traffic Safety Management and Control (QXT 729-2024), and Table 2 is explicitly defined with reference to ISO 8348 (Lines 168–202; Reference 33, Lines 909–910).

（3）The weather state in Table 4 is difficult to understand, please provide more details on that.

Respond: Thank you. We agree that the coding scheme required clearer explanation. In the revised manuscript, we added an explicit clarification in Section 2.1.3 to explain that the first index denotes the weather category/severity level and the second index denotes the road-slipperiness level. We also clarified the coding rule again in Section 2.2 and Table 4 (Lines 219–224; Lines 247–252).

（4）Table 5 must be improved.

Respond: Thank you. We revised Table 5 to improve readability and hierarchical clarity while keeping the original indicator system and evaluation logic unchanged. The revised presentation makes the target–function–factor–indicator structure clearer and easier to follow (Lines 275–289).

（5）I don't see the purpose of section 2.3.2, please revise and provide more physical details on that.

Respond: Thank you. We revised the opening of Section 2.3.2 to clarify its methodological purpose. The revised text now states more explicitly that weight determination is introduced because the relative importance of different indicators cannot be represented adequately by either expert judgment alone or sample dispersion alone, and that the combined-weight procedure is used to preserve both engineering interpretability and data-based differentiation before fuzzy aggregation (Lines 291–326).

（6）Please provide more insights about SCANeR Studio software.

Respond: Thank you. We strengthened the description of SCANeR Studio in Section 3.2. The revised manuscript now describes SCANeR not only as a visualization environment but also as the core simulation platform for vehicle dynamics modeling, virtual sensor configuration, scenario editing, and real-time interaction data generation under controlled weather–road perturbation conditions (Lines 389–413).

（7）More details about the AEB system used in this work are required. Simulation results depend on the AEB structure, so this is necessary for reproducibility.

Respond: Thank you. We agree that simulation results depend on the AEB structure and revised Section 3.3 accordingly. The revised manuscript now clarifies that the TTC-based controller is used as a simplified comparative AEB model rather than a full production-grade braking system, and explains its role, inputs, trigger logic, and intended use in controlled comparative evaluation (Lines 421–450).

（8）Maybe the TTC-based AEB should be presented as part of the methodology, and not straight into the simulation section.

Respond: Thank you for this helpful suggestion. We agree with the Reviewer’s point and revised the manuscript accordingly. The revised Section 3.3 now states explicitly that, although implemented within the simulation environment, the TTC-based AEB logic constitutes part of the methodological setup of the proposed evaluation framework because it defines how environmental perturbations are translated into trigger-response behavior and braking-performance outcomes (Lines 421–424).

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Good work; The revised paper has been enhanced with the input given by the reviewers.

Author Response

Comment:

Good work; The revised paper has been enhanced with the input given by the reviewers.

Response: Thank you very much for your positive and encouraging comments. We sincerely appreciate your recognition of our revisions and are grateful for the valuable suggestions provided by the reviewers, which have helped us improve the quality of the manuscript.

Author Response File: Author Response.doc

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for your detailed revisions and thoughtful responses to my comments. The corrections have strengthened the manuscript, and I find it suitable for publication.

Author Response

Comment:

Thank you for your detailed revisions and thoughtful responses to my comments. The corrections have strengthened the manuscript, and I find it suitable for publication.

Response: We sincerely thank you for your positive evaluation of our revised manuscript. We greatly appreciate your thoughtful comments and recognition of the revisions we have made. Your valuable feedback has significantly contributed to improving the quality and presentation of the paper. We are grateful that you consider the manuscript suitable for publication.

Author Response File: Author Response.doc

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript has been improved and can be accepted now

Author Response

Comment:

The manuscript has been improved and can be accepted now.

Response: Thank you very much for your positive comments. We sincerely appreciate your recognition of the improvements made to the manuscript. We are grateful for your valuable feedback throughout the review process, which has helped us strengthen the paper.

Author Response File: Author Response.doc