Synthesis of a Path-Planning Algorithm for Autonomous Robots Moving in a Game Environment during Collision Avoidance

Round 1

Reviewer 1 Report

Overview:
This paper proposes a safe mobile robot control process in collision situations using the model of a multistep matrix game of many participants in the form of a dual linear programming problem and provides a method to perform optimization on the process. The authors state this is a novel contribution because the process is robust in the face of uncertainty with respect to incomplete cooperation between mobile robots.

Major comments:
(1) The study would be improved by describing the actionable outcomes of the performed analysis further. The need for safe mobile robot control process in collision situations in the face of uncertainty is referenced with a somewhat awkwardly worded sentence:

This justifies the problem analysis and method design of safe mobile robot control using game theory and artificial intelligence methods what works show Spica et al. [19], Li and Vorobeychik [20], Liu et al. [21].

The sentence above should be reworded to read more clearly and make it obvious to readers exactly what problem statement the community has defined and how it is tied to game theory and artificial intelligence.

However, even with if the above sentence is more clearly worded it is still unclear what the applied instances of this problem statement in the wild are. Providing and example and describing why the authors applied approach is particularly actionable would improve the impact of the paper, help make it more applicable to the readership of Electronics and result in more citations in the future.

In particular, referencing the similar need in the stochastic simulation community with respect to identifying the location of faults/bugs in the face of uncertainty, may help readers appreciate the impact and applicability of the problem and how solutions using artificial intelligence approaches like fuzzy logic are an effective approach to addressing these situations.

Gore, Ross, et al. "Statistical debugging for simulations." ACM Transactions on Modeling and Computer Simulation (TOMACS) 25.3 (2015): 1-26.

Kamensky, David, Ross Gore, and Paul F. Reynolds. "Applying enhanced fault localization technology to Monte Carlo simulations." Proceedings of the 2011 Winter Simulation Conference (WSC). IEEE, 2011.

Hao, P., et al. "Statistical fault localization in decision support system based on probability distribution criterion." 2013 Joint IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS). IEEE, 2013.

(2) The rationale and methods/design related to the experiments for which results are presented in section 3 is limited. Expanding this description specifically with respect to  the variables and their bindings which control traffic (K), weather (d_s) and visibility (also d_s?) would improve the paper. In particular, helping the reader understand why and what values bound to these variables represent, and why the range of values tested in the experiments reflects an effective set of conditions to generate results upon which conclusions can be made.

(3) The paper would be improved if an analysis / discussion was included with respect to identified validity threats. In particular, analysis/discussion of threats to internal, external and construct validity should be provided to readers so the results and conclusion of the study can be placed in appropriate context.

Parker, Randall M. "Threats to the validity of research." Rehabilitation Counseling Bulletin (1993).

(4) In the replication crisis era, the programs written in the Matlab/Simulink software along with the raw data and scripts used to compute and plot the measures shown in the figures should be provided to the reviewers and readers. In addition the source code used to replicate results (tables and figures) the authors report in the paper should also be provided to the reviewers.

Minor comments
(4) A number of the figures (4, 9 and 13) all employ color palettes which will be difficult to read for colorblind readers. As many as 8% of men and 0.5% of women are affected with the common form of red-green color blindness that creates no contrast between the two colors. The figures would be improved if a different color choice is used. There are numerous options that offer superior readability at: https://davidmathlogic.com/colorblind/#%23D81B60-%231E88E5-%23FFC107-%23004D40. In addition of a legend as well as increasing the thickness of the lines and the size of the data points would improve readability of Figures 6-13. Finally, Figures 2 and 3 are not centered. It is unclear why this choice is made for these figures and not others. The lack of consistency is confusing for readers and should be corrected.

(5) The Section number "2" is repeated twice. Line 62 and 111. I assume this is a typo, configuration error with the word processing software used. It is confusing for readers and should be corrected.

(6) The size of font for the text changes on lines 192 and 193 in the sentence, "Figure 13 shows a comparison of the autonomous surface safe ship path when passing other encountered autonomous surface ships, ..." After the comma the size of the text reverts back to its original size. Again, I assume this is a typo, configuration error with the word processing software used. It is confusing for readers and should be corrected.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This is a well considered piece of research. I have no question on the viability or justification of the methods and results. However, there are some readability issues in certain places (the abstract for example) that said, I believe that this is moderate and easily fixed. 

One thing that I am particularly impressed with is the use of variable dynamics to emulate real-world conditions as this is often left out of research that employs simulation software to assess robotic control systems.

Interesting and appropriate use of game theory / optimization algorithms to solve the problem of collision avoidance/detection/safest route in real world scenarios.

It may be beneficial for the reader to see a future work section? or at least some idea of what the next steps of this research are in the conclusion? for example a practical assessment of the gamification / optimization model using real mobile robots? 

I understand this may or may not be research carried out by yourself, but it would be significant for future research in this area for you to detail what the potential applications of this research could entail. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments to the authors

Manuscript ID: electronics-1139397

Title: Synthesis of Path Planning Algorithm of Autonomous Robots Moving in a Game Environment During Collision Avoidance

1) There are many English issues that need to be corrected. Just to name a few:

a) “… synthesize an [a] game algorithms …”

b) “According [to] Bist …”

Please make sure that all grammar and English issues are corrected in the revised version of the manuscript.

2) What is the contribution(s) of the manuscript. States clearly in Section 1.

3) Provide a brief outline of the paper at the end of Section 1.

4) More details about optimization problem (7) are required. Is it a convex problem? How to solve it in real-time? Note that one possible way to solve the optimization problem in real-time is to use the continuous-time parallel solvers. Please discuss this in the manuscript. See 10.1109/LCSYS.2018.2889963 and 10.1109/TAC.2018.2867359 for more details.

5) Discussion section is too brief and needs to be enhanced. It should include all observations from the results.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

No further comments.