The Effect of Rainfall and Illumination on Automotive Sensors Detection Performance

Round 1

Reviewer 1 Report

Review:

The Effect of Rainfall and Illumination on Automotive Sensors Detection Performance

Overview:

This article shows us an investigation into how data gathered from sensors like cameras, radar, and lidar in some typical unfavorable environmental conditions can help prevent traffic accidents. All these sensors are the tools for improving the Advanced Driver Assistance System (ADAS), which was mainly mentioned in this work.

Main points in this article:

In the introduction, the authors describe the causes of traffic accidents as well as the application of ADAS to prevent these terrible incidents. Humans, weather, and cars are responsible for these issues. The correct application of ADAS sensors (cameras, radar, and LiDAR) also contributes to enhanced driving safety. By focusing on the two most basic states, day and night, the authors describe several sensor accuracy test cases and variations during these processes. Driving at night under poor lighting and in the rain was shown to be the most dangerous, resulting in the highest number of accidents, fatalities, and injuries. It can be seen that each type of sensor has distinct strengths and weaknesses in different situations, hence it is vital to analyze the effect of the surrounding environment on these sensors. Other studies on this subject have mainly been conducted under artificial conditions, i.e., indoors, therefore it is probable that there will be errors when it is applied in the real world. All the experiments in this work were conducted in natural outside conditions, which facilitates practical applicability. The authors have also included brief summaries of the sections featured in this work at the conclusion of the introduction.

Section 2, test facilities, is comprised of three sections: test track, tested sensors, and ground truth definition. These sections introduce the specifications of the sensors and how to establish a standard sensor (RS-reference), which serves as the sensor's basis in this article. The third section, entitled "Test Methodology," offers the reader information on how the authors altered the parameters of the weather conditions and the maneuvers that may occur in real-world situations in order to examine the impacts of rainfall and illumination. Sections 4 and 5 mainly examine, explain, and discuss the statistics from real sensor measurement and evaluation for the most important vehicle sensors, indicating the sensor limitations for ADAS. And the last section, Section 6, is conclusions and some suggestions for future improvements by the authors.

Recommendations:

- there are some misspellings: gray level that lack of “–“ in line 46, “Tehrefor” in line 193, and last “meahile” in line 335.

- In this article, there are still some spelling errors that can be pointed out, such as not capitalizing after punctuation (line 334) or capitalization after commas (lines 37,48) and misspelling (line). 193 - therefore and meanwhile in line335)

- The term "Lidar" should be uniformly written throughout the article, because in the first part, the authors write "lidar" while the second part uses "LiDAR"

- The final paragraph of the introduction summarizes the sections that will be included in the paper, but it would be more appropriate for the writers to include these summaries in the first sentence of each section so that the reader can more easily follow up.

- Section 2.3 raises the question of whether the RS-reference error should be further calculated. Since this is another instrument utilized in this investigation.

- Why do the authors say (line 184) there are "nine different maneuvers in total" when, in Figures 4 and 3, only 7 cases are listed?

- Section 6, the authors' conclusion without exact numbers from the preceding data, requires extra consideration. In addition, the question arises as to whether we may integrate these sensors for simultaneous operation, hence enhancing the abilities of ADAS.

Conclusion:

The authors have collected a lot of data and analyzed it in a reasonable way. This has led to work that is both complete and useful for preventing the common traffic accidents that happen today. Moreover, the authors should also consider and correct certain problems in this article.

Author Response

Dear Reviewer,

We would like to thank you for your careful and thorough reading of this manuscript and for the thoughtful comments and constructive suggestions, which have contributed significantly to the quality of this manuscript.

Please see the attachment for detailed revision information

Author Response File: Author Response.docx

Reviewer 2 Report

·        In the concluding section of the paper, where they summarize their findings, the authors state that "Our analysis focused on lateral distance detection, and we evaluated the experimental results quantitatively." However, neither the summary nor the conclusions present any quantitative conclusions about the paper's findings.

·        The prepared paper should include a scientific study and its findings. It cannot be used and evaluated as a promotional brochure for a test track (i.e., DigiTrans) used on paper. The authors should simplify the paper in this respect. Especially section 2.1.

·        The references should be provided for the technical specifications of the test sensors given in Table 1. After all, this equipment was not developed by the authors, and manufacturer information was used.

·        In the paper, the material and method parts and the findings and discussion are intertwined. a confusing arrangement. Eq. 1–6 are the findings of the paper, or are they the equations used to determine the findings of the paper? Why are these equations not given in section 2.2?

·        Deep and mechanistic discussions are required to explain the results obtained. The section "Results and discussion" is reportorial. The study's comparative analysis with others must be thoroughly discussed. A few examples of recent studies for comparison and reference: https://doi.org/10.1177/0954407017690962, https://doi.org/10.1049/iet-its.2019.0322.

·        In the "structure of weather combinations" in Figure 4, there are heavy rain conditions for the night situation. Also, the "Test Track" used seems to have this feature as well. Well, why are there no test results for heavy rain conditions at night among the results of the paper? Will the authors have an explanation for this? (In the range of Tables 2–7.)

·        It is commendable to present the predictions for future studies in the conclusion section. However, the conclusions section should mostly contain the results of the current paper. The conclusion part should be rewritten as discussing the information containing the numerical results of the paper.

·        Although the paper contains important findings regarding the faults of the sensors used in the vehicle for different weather conditions, the results obtained should be expressed numerically and should provide conclusive evidence and recommendations to the reader and the relevant literature.

Author Response

Dear reviewer,

We would like to express our gratitude for taking the time to review our manuscript titled "The Effect of Rainfall and Illumination on Automotive Sensors Detection Performance". We appreciate your constructive feedback and suggestions to improve the quality of our work. We have carefully considered all of your comments, and in response, we have revised the manuscript accordingly.

Firstly, we have revised the concluding section of the paper to include quantitative conclusions about our findings. We have also added a more detailed discussion and analysis of our results in the "Results and Discussion" section, and have provided a comparative analysis with other recent studies to further support our findings. Additionally, we have revised the conclusion section of the paper to provide conclusive evidence and recommendations, which are supported by the quantitative numerical results from our measurement.

The change details please find the revision comments file in the attachment.

Thanks a lot!

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The final version of the article is suitable for printing. You can include the following articles in the reference section as well.

https://doi.org/10.1177/0954407017690962,

https://doi.org/10.1049/iet-its.2019.0322.