Objectively Scoring the Human-Likeness of Artificial Driver Models

Round 1

Reviewer 1 Report

In this manuscript, the authors proposed an attempt to make the plausibility and human-likeness of vehicles’ trajectories objectively measurable by multidimensional quality function. This research is interesting and has practical significance. For this manuscript, I still have some suggestions and doubts. And my first instinct for this manuscript is a minor revision. Some of the suggestions and questions are listed below.

1. In the Introduction, the arrangement of the whole manuscript should be presented, for example: In Section 2, the model of …; In Section 3, ….

2. The fonts in Table 1 are not uniform, some are italic, and some are not. Please check the format of the whole manuscript carefully.

3. The reason to choose the parameters for distinguishing scenarios in Table 2 should be mentioned in the manuscript.

4. How is the priority order of the context parameters determined in Table 3?

5. The notation “w_i” in Eq.(1) should be interpreted in the manuscript.

6. In section 3.1, the specific process of data creation should be disclosed in the manuscript.

7. In section 3.1, the calculation process of behavior and context parameters should be disclosed.

8. I suggest that the contents in Figure 5 should be described in more detail.

9. In Figure 7, the label of the Y-axis should be presented in the manuscript.

Author Response

Dear Reviewer,

please find my comment in the attached file. Could you please clarify your feedback on point 7 as described in the attachment?

 

Author Response File: Author Response.docx

Reviewer 2 Report

The paper presented survey based analysis of human feedback of vehicles’ trajectories to support urban traffic. The study is based on survey which is not supported by any validation that could be effectively used for design and operation. I suggest to use simulation, and measurable parameters and related to human behavior models, driving models, and performance measures.

1. What is the main question addressed by the research? The paper presented quality function based to measure human-likeness of vehicles’ trajectories. And compare performance parameters of human-like driving behavior in similar situations. 2. Do you consider the topic original or relevant in the field? Does it address a specific gap in the field? The topic is important, but the approach is based on surveys which will not provide concrete results to build reasonable controls for possible implementations. 3. What does it add to the subject area compared with other published material? It provides analysis of human-likeness of vehicles, which is personalized and not linked to some technical aspects of the driving behavior. 4. What specific improvements should the authors consider regarding the methodology? What further controls should be considered? The authors should include more analysis. I suggest to use simulation, and measurable parameters and related to human behavior models, driving models, and performance measures. 5. Are the conclusions consistent with the evidence and arguments presented and do they address the main question posed? Yes, but general and not helpful to get evidences of the analysis without simulation and link to technical aspects of the driving. 6. Are the references appropriate? Yes, but include more about simulation and quantitative analysis of the driving trajectories. 7. Please include any additional comments on the tables and figures. The tables and figures should include simulation and analysis of quantitative measures of driving trajectories, specify constraints, performance measures.

Author Response

Dear Reviewer,

could you please clarify your feedback on certain aspects as described in the attachment?

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

the revisions are fine