Smart Stick Navigation System for Visually Impaired Based on Machine Learning Algorithms Using Sensors Data

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the authors present a Smart Stick Navigation System for Visually Impaired Based on Machine Learning Algorithms Using Sensors Data.

The text is well-written and organized.

The strengths and suggestions are:

Abstract:

Strengths: Provides a clear overview of the system's components and the machine learning algorithms used.

Suggestions for Improvement:
The abstract could be more concise. Consider summarizing the key findings more succinctly.
Include a brief mention of the practical implications or potential real-world impact.

Introduction:

Strengths: Clearly outlines the challenges faced by visually impaired people and the need for advanced assistive technologies.

Suggestions for Improvement:
The introduction is quite long. Consider condensing some of the background information to focus more on the current study's motivation and objectives.
Add a brief summary of the contributions at the end of the introduction.

Literature Review:

Strengths: Provides a comprehensive review of related works, highlighting the advancements and limitations of existing systems.

Suggestions for Improvement:
Some references are outdated. Including more recent studies would strengthen the literature review.
A table summarizing the key features and performance metrics of related works could make the comparisons clearer.

Methodology:

Strengths: Detailed description of the system design, hardware components, and machine learning algorithms used.

Suggestions for Improvement:
The methodology section is very detailed but could benefit from clearer subsections. For example, separate the descriptions of hardware, software, and machine learning processes more distinctly.
Include diagrams for the flow of data between sensors, the microcontroller, and the output devices to improve clarity.

Results:

Strengths: Thorough presentation of experimental results with clear figures and tables.

Suggestions for Improvement:
The results section is quite technical. Consider adding more interpretation of the results in layman's terms to enhance readability.
Highlight the most significant findings in a brief summary at the end of this section.

Discussion:

Strengths: Provides a good analysis of the results, comparing them with related works.

Suggestions for Improvement:
The discussion could be expanded to include more on the practical implications and potential limitations of the study.
Consider discussing the potential for future work in more detail, including how the system could be improved or extended.

Conclusion:

Strengths: Summarizes the main findings and the effectiveness of the proposed system.

Suggestions for Improvement:
The conclusion is a bit brief. It could include a more detailed discussion on the impact of the findings and the next steps for research.

References:

Strengths: References are relevant to the study.

Suggestions for Improvement:
Ensure all references are up-to-date and formatted consistently.
Consider adding more recent references to strengthen the review.

General Comments:

The paper is well-structured but could benefit from a more concise and focused narrative.

Ensure consistency in formatting, especially in figures, tables, and references.
Consider adding more visual aids, like flowcharts and summary tables, to enhance understanding.

Comments on the Quality of English Language

See above.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper proposes the design and implementation of a smart stick for visually impaired people using an Arduino Nano microcontroller, multiple sensors and actuators, backed by machine learning algorithms. The paper is interesting. However, some issues need to be solved:

1.       English needs polishing. Some examples: a) lines 11-12: the sentence “To address this, a researcher developed a smart stick as a more effective aid.” is confusing (“a researcher developed a smart stick” may be replaced by “researchers have developed smart sticks”; also, the sequence “as a more effective aid” refers to what? (“more effective” than what?)); b) line 70:  what “negotiating” means in the context? c) lines 204-205: the sequence “offering crucial insights into the health and ecological popularity of the VIP” looks strange. Please read carefully the entire manuscript and make the necessary changes.

2.       The list of keywords needs to be reshaped: I suggest visually impaired, smart cane, Arduino Nano microcontroller; multisensor system; machine learning.

3.       Lines 73-74: what “detection variety” means in the context? Maybe “detection range” is more appropriate.

4.       The abbreviation SS needs to be fully spelled the first time it is mentioned.

5.       What is the timestep used in the algorithms? On what basis it was selected? for example, what is the timestep used to acquire data from the sensors?

6.       The ML-related part of the paper is confusing since much information is missing: a) how the dataset to train/test the ML algorithms was obtained? Does this dataset cover the entire spectrum of movements/scenarios encountered by visually impaired people? b) Figure 4 represents the way the ML algorithms are trained/tested, but how the trained model is used in real-life scenarios? c) the testing dataset seems to have only a few records (i.e., 98 as per Table 3 - is it enough?); d) the parameters for each ML algorithm (e.g., for NN: number of layers, number of neurons per layer, activation function type, etc.) need to be presented. Please provide all the implementation details.

7.       Table 6: the column name “Algorithm” is wrong. For example, CNN, Lidar data, or YOLO are not algorithms.

Comments on the Quality of English Language

English needs polishing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript is interesting from a scientific point of view and well structured.

Here are some requests to improve it further:

1) At the end of Section 1, "Introduction", authors must report the structure of the manuscript, with a brief description of the content of the subsequent sections

2) In Section 2.2, a more detailed description of the real connections of the electronic modules shown in Figure 2 must be reported with an additional image.

3) In Section 3, more details on the ten ML algorithms must be provided, also at the code level, to illustrate the most important characteristic aspects of the various models and the relative differences in obtaining the quantities and parameters of interest, then reported in the next section.

4) In Section 6, before comparing with the scientific literature, the authors must summarize the results obtained, highlighting the strengths and innovative aspects of the research.

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have successfully solved all my comments and concerns.

Comments on the Quality of English Language

English needs minor polishing.

Author Response

Thank you very much for taking the time to review our manuscript. We appreciate the reviewer's comments, which have improved our paper.

Reviewer 3 Report

Comments and Suggestions for Authors

After the careful review by the authors, the article can be accepted for publication in its current form

Author Response

Thank you very much for taking the time to review our manuscript. We appreciate the reviewer's comments, which have improved our paper.