ROS Implementation of Planning and Robust Control Strategies for Autonomous Vehicles

Round 1

Reviewer 1 Report

The paper presents an autonomous vehicle architecture using ROS2 framework and advanced control algorithms. Lateral vehicle dynamics controllers are designed and tested on an RC car platform. The objective is to achieve optimal tracking and stability using LTI/H∞ and LPV/H∞ approaches, with both simulation and experimental results presented.

However, some problems still need to be improved before a possible publication. Specifically,

1) In the introduction part of the paper, it is recommended to expand the literature review to provide a more comprehensive overview of existing research and state-of-the-art developments related to control strategies for autonomous vehicles: A systematic survey of control techniques and applications in connected and automated vehicles. For the extreme conditions, such as: Dynamic drifting control for general path tracking of autonomous vehicles. This would help highlight the novelty and unique contributions of the research gap. So, more paper should be reviewed.

2) The labels of all figures keep the same, for example, in section 3.3.2, there is a figure named Figure 3.3.2, please check them, which makes the paper more readable.

3) How is the RC car platform in GiPSA-LAB used to facilitate the development and testing of the lateral vehicle dynamics controllers? It is better to give a brief explanation. Since in Section 2.1, there is the general hierachical framework for AVs, so it is better to explain how your algorithm is applied to this platrotm.

4) Based on the findings of the paper, what are the potential implications for the future development and improvement of autonomous vehicle architectures and control algorithms?

5) In conclusion, summarizing the main findings and results into concise bullet points enhances the comprehensiveness of this paper according to your main results and findings.

 

 

 

Minor

 

Author Response

The notes are included in the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper is well structured from methodological side, even though the system identification is not detailed enough. On some points the text becomes too brief and omitting important details. So the conclusion is the paper can be accepted after minor revisions on the style and presentation. However, note that number of corrections is not small. 

1. Not sure is it better to have expanded the ROS abbreviation in the title.

2. Introduction contains enough information. Maybe a little more review onto various kinds of mathematical models employed in vehicle steering would be beneficial, since the design of the controller is primarily dependent on the selected model.

3. Don't see a clear point of subsection 1.1 in the introduction. It is natural to conclude introduction with a contribution statement, no need to separate in a subsection.

4. Please add a short paragraph about further organization of the paper in the end of the introduction.

5. The quality of the figure with Architecture could be improved. Some texts are not well readable. Some brief comments what is the implementation of each block in the figure would be interesting.

6. Rendering quality of figure 6, 7, 11, 12 is not enough too.

7. Figure 3.3.2 doesn't follow a consistent numbering.

8. Equation (8) is not properly aligned in the text.

9. Some justification is needed why bicycle model is chosen instead of four wheel model.

10. That is important one ! Not enough information what exact identification method is selected, how the experimental data is chosen, what are the confidence bounds for the parameters, what is the level of fit between experimental data and the model given a different set of experimental data. All the typical things for system identification.

11. Conclusion and discussion of the results is too brief. Extend with additional analysis.

P.S.: I'll keep my right to add more notes in addition to the above in the following revisions as further read the paper

English is grammatically correct, however the text may become a bit unclear around abbreviations.

Author Response

The notes are included in the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

The main emphasis of this project is centered on constructing an autonomous vehicle architecture by leveraging the ROS2 framework and incorporating advanced control algorithms. To facilitate the development and implementation of lateral vehicle dynamics controllers, an experimental platform using an RC car from GiPSA-LAB is employed. The ultimate goal is to design sturdy controllers capable of achieving optimal tracking and stability. To address this challenge, two approaches, namely LTI/H∞ and LPV/H∞, were utilized, demonstrating simulation and experimental outcomes.

In general, according to the reviewer's assessment, the paper introduces a captivating concept and some solid research outcomes that fit coherently within the journal's theme. However, the reviewer has specific comments that require thoughtful consideration and should be duly addressed. 

1. Considering that the chosen methodology for vehicle path-tracking is linear robust control, it becomes essential to conduct a more comprehensive review of the relevant literature. The current reference list in this area seems outdated and does not include state-of-the-art publications from reputable journals within the field. To address this, the authors should thoroughly review recently published papers like [R1] https://doi.org/10.1109/TITS.2022.3207023, [R2] https://doi.org/10.1115/1.4051466, [R3] https://doi.org/10.1109/TMECH.2022.3146727, and other relevant sources to strengthen the paper's foundation.

2. Further elaboration and additional details are required to explain the design of the performance weighting transfer function. It remains unclear whether the implementation follows the Glover-McFarlane method or the Mixed-sensitivity approach. Clarifying this aspect in the paper is essential to enhance the reader's understanding of the methodology employed.

3. The language used in this paper requires further refinement and polishing.

The language used in this paper requires further refinement and polishing.

Author Response

The notes are included in the attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

To the reviewer's read, the revision looks good and the paper can be accepted as is.